CN109447048B

CN109447048B - Artificial intelligence early warning system

Info

Publication number: CN109447048B
Application number: CN201811616752.5A
Authority: CN
Inventors: 詹志超
Original assignee: Suzhou Shanchi Numerical Control System Integration Co ltd
Current assignee: Suzhou Shanchi Numerical Control System Integration Co ltd
Priority date: 2018-12-25
Filing date: 2018-12-25
Publication date: 2020-12-25
Anticipated expiration: 2038-12-25
Also published as: CN109447048A

Abstract

The invention relates to an artificial intelligence early warning system which comprises an intelligent internet of things and risk factor data acquisition system (100), a risk factor management system (200), cloud computing (300), cloud storage (400), a cloud database (500), an artificial intelligence early warning operating system (600), an artificial intelligence early warning server (700), an internet + distributed early warning kiosk (800), a five-level artificial intelligence early warning system (900), a four-level artificial intelligence early warning system (1000), a three-level artificial intelligence early warning system (1100), a two-level artificial intelligence early warning system (1200) and a one-level artificial intelligence early warning system (1300). According to the invention, an artificial intelligent early warning system is used for collecting, contrastively analyzing, reasoning, evaluating, cloud computing, cloud storage, grading alarm and coping prevention and control on risk factors; the all-weather 24-hour monitoring on the peripheral control points of the police kiosk is realized, the information sharing can be realized for users, the utilization rate of information resources is improved, and the safety guarantee is increased for maintaining the frontier stability.

Description

Artificial intelligence early warning system

Technical Field

The invention relates to the field of intelligent security early warning, in particular to an artificial intelligent early warning system for security stability maintenance.

Background

The artificial intelligence early warning system is a comprehensive intelligent early warning system which is established by effectively integrating an advanced information technology, a data communication transmission technology, an electronic sensing technology, an electronic control technology, an AI computer early warning processing technology, an AI artificial intelligence early warning operation technology, a risk factor acquisition technology, a risk factor identification technology, a big data analysis technology, a cloud computing technology, a cloud storage technology, a cloud database technology, an Internet + distributed early warning kiosk and the like on the whole artificial intelligence early warning system in a large range and in an all-around manner, and is real-time, accurate and efficient.

Along with the normalization of the maintenance measures of the western West of China, the West of China keeps a stable situation for a long time, the rapid development of economy is guaranteed, and the artificial intelligent early warning system lays a good foundation for fundamentally solving the deep-level problem affecting long-term safety.

Disclosure of Invention

The invention provides an artificial intelligent early warning system for overcoming the problems of a security monitoring blind area, inadequate early warning measure, untimely precaution and the like in the existing security system.

For realize above-mentioned use artificial intelligence early warning system carry out face identification information, speech recognition information, dynamic identification information, biological feature identification information, infrared pronunciation detection information, infrared night vision detection information, early warning unmanned aerial vehicle cruise information and thing networking intelligent security information's collection, carry out contrastive analysis, sign reasoning, early warning value aassessment, risk factor reply and the hierarchical early warning's of risk management evaluation purpose again to the risk factor source, the utility model provides a following technical scheme: the utility model provides an artificial intelligence early warning system, including intelligent thing allies oneself with and risk factor data acquisition system 100, risk factor management system 200, cloud computing 300, cloud storage 400, cloud database 500, artificial intelligence early warning operating system 600, artificial intelligence early warning server 700, internet + distributed early warning pavilion 800, five-stage artificial intelligence early warning system 900, level four artificial intelligence early warning system 1000, level three artificial intelligence early warning system 1100, level two artificial intelligence early warning system 1200, level one artificial intelligence early warning system 1300. According to the invention, the artificial intelligent early warning system is used for collecting, contrastively analyzing, reasoning, evaluating, cloud computing, cloud storage, grading alarm and coping prevention and control on risk factors, all-weather 24-hour monitoring on the peripheral control points of the police kiosk is realized, information sharing can be realized by a user, the utilization rate of information resources is improved, and safety guarantee is increased for maintaining frontier stability.

The invention provides an artificial intelligence early warning system, which comprises an intelligent Internet of things and risk factor data acquisition system: the system is used for acquiring a face recognition system, a voice recognition system, a dynamic recognition system, a biological feature recognition system, an infrared voice detection system, an infrared night vision detection system, an early warning unmanned aerial vehicle cruise system and an intelligent Internet of things security monitoring system, and transmitting the acquired information to a risk factor management system.

Risk factor management system 200: the intelligent Internet of things and risk factor data acquisition system is used for receiving the intelligent Internet of things and risk factor data acquisition system, classifying the acquired face identification information, voice identification information, dynamic identification information, biological characteristic identification information, infrared voice detection information, infrared night vision detection information, early warning unmanned aerial vehicle cruise information and Internet of things intelligent security information, and transmitting the classified image characteristic information and voice information to the risk factor identification module, wherein the image characteristic information comprises acquired face images, dynamic characteristic information and biological characteristic information; the voice information comprises collected voice information, the original image feature information comprises a face image blacklist, a dynamic feature information blacklist and a biological feature information blacklist which are stored in a storage unit, and the original voice information comprises a voice information blacklist which is stored in the storage unit.

The risk factor recognition module performs comparative analysis, hierarchical reasoning and early warning value evaluation on the mass data information through cloud computing, and then returns the processing result to the user and performs cloud storage 400.

Risk factor identification module 220: the early warning system is used for comparing and analyzing the image characteristic information with the original image characteristic information to judge whether the image characteristic information is matched with the information of the image model base or not, if the image characteristic information is judged to be 'Yes', early warning prompt information is generated, and then the signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module 250 is generated, the early warning level warning information is regenerated into early warning signals to be fed back to a primary early warning system, namely a risk management evaluation module 260, and the prompt information is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program is judged to be 'No', then an early warning value No evaluation result is transmitted to an image information database, namely a cloud, if the comparison analysis with the original image characteristic information is judged to be "No", the signal is transmitted to a daily activity sign inference program, namely a risk factor inference module 230.

Risk factor inference module 230: the daily activity sign reasoning program generates early warning prompt information if judging to be 'Yes', generates early warning prompt information and then transmits the signal to an early warning value evaluation program, namely a risk factor evaluation module 240, the early warning value evaluation program generates early warning level warning information, namely a risk factor response module 250 if judging to be 'Yes', the early warning level warning information is generated into early warning signals and then fed back to a primary early warning system, namely a risk management evaluation module 260, and transmits the prompt information to an artificial intelligent early warning operation system 600, the early warning value evaluation program generates early warning level No evaluation result if judging to be 'No', the early warning value No evaluation result is transmitted to an image information database, namely a cloud database 500, and the daily activity sign reasoning program transmits the signal to a weekly activity sign reasoning program, namely a risk factor reasoning module 230 if judging to be 'No', the acquired image characteristic information is subjected to activity evidence cycle reasoning on daily, weekly, monthly, quarterly and yearly.

Risk factor identification module 220: the voice information and the original voice information are compared, analyzed and judged to judge whether the voice information is matched with the information of the voice model base or not, if the voice information is judged to be 'Yes', early warning prompt information is generated, and then the signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module 250 is generated, the early warning level warning information is regenerated into early warning signals and is fed back to an upper early warning system, namely a risk management evaluation module 260, and the prompt information is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program is judged to be 'No', then an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database 500, if the voice information is compared with the original voice information and analyzed, and a No is judged, the signal is transmitted to a keyword recognition inference program, namely a risk factor inference module 230.

Risk factor inference module 230: an inference program for keyword recognition, continuous speech recognition, a grammar analyzer and semantic information analysis, wherein if the keyword recognition inference program judges "Yes", the keyword recognition inference program generates early warning prompt information, generates early warning prompt information and then transmits signals to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program judges "Yes", the early warning value evaluation program generates early warning level warning information, namely a risk factor response module 250, the early warning level warning information is regenerated into early warning signals and is fed back to an upper early warning system, namely a risk management evaluation module 260, and transmits the prompt information to an artificial intelligent early warning operation system 600, if the early warning value evaluation program judges "No", the early warning value No evaluation result is transmitted to a speech information database, namely a cloud database 500, and if the keyword recognition inference program judges "No", the signals are transmitted to a continuous speech recognition inference program, namely a risk factor inference module 230, the steps are used for carrying out the cycle reasoning of keyword recognition, continuous voice recognition, a grammar analyzer and a semantic information analysis reasoning program on the collected voice information.

Risk factor assessment module 240: and is configured to evaluate the early warning information generated by the risk factor inference module, and transmit the prompt information to the risk factor response module 250.

Risk factor handling module 250: and is used for generating an early warning signal and feeding the early warning signal back to the upper-level early warning system, namely the risk management evaluation module 260, according to the measures taken by the early warning level alarm information generated by the risk factor evaluation module.

Risk management evaluation module 260: the system is used for evaluating and managing the early warning information fed back to the upper level by the risk factor handling module, receiving the operation instruction sent by the upper level early warning system, and replying the execution steps, processes, measures and implementation results.

The cloud computing 300: the design is carried out based on an open-source Hadoop architecture, a user can develop a distributed program under the condition that distributed bottom-layer details are not known, high-speed operation and storage are carried out by utilizing cluster advantages, and the cloud computing 300 is used for carrying out comparative analysis, hierarchical reasoning and early warning value evaluation on the risk factor identification module 220, the risk factor reasoning module 230 and the risk factor evaluation module 240 on the computing distributed computer and mass data, and then returns a processing result to the user and carries out cloud storage 400.

Cloud storage 400: the cluster application for the cloud computing system integrates the distributed system files through application software to cooperatively work so as to provide data storage and service access for users, the system is used for comparing the collected face image, dynamic characteristic information, biological characteristic information and voice information with the face image blacklist, the dynamic characteristic information blacklist, the biological characteristic information blacklist and the voice information blacklist in the storage module by arranging an online data storage module, wherein the face image blacklist, the dynamic characteristic information blacklist, the biological characteristic information blacklist and the voice information blacklist are stored in the storage module, if the similarity reaches a preset early warning value, the early warning system generates early warning prompt information for risk factor reasoning and evaluation in time, generates early warning level warning information and feeds the early warning level warning information back to the previous early warning system for risk management and evaluation.

Cloud database 500: the cloud database 500 is divided into an original voice information database, an original image characteristic information database, an original information database, a real-time risk factor acquisition image information database, a real-time risk factor acquisition voice information database, a risk factor identification database, a risk factor reasoning database, a risk factor evaluation database, a risk factor response database, a risk factor management evaluation database, a real-time judgment basis database, a judgment rule database and an accident case database, the cloud database 500 is used for storing risk factor information data acquired by the intelligent physical association and risk factor data acquisition system 100, the system is used for storing voice information contrastive analysis, keyword recognition, continuous voice recognition, a grammar analyzer, semantic information analysis and image characteristic information contrastive analysis, daily activity sign reasoning, weekly activity sign reasoning, monthly activity sign reasoning, quarterly activity sign reasoning, annual activity sign reasoning, early warning prompt information generation, early warning value evaluation, early warning level warning information generation, early warning signal feedback to a previous early warning system, information data which is updated in real time according to data transmitted by a cloud computing 300 through a cloud storage 400, information data which is used for storing prompt information, grading information, early warning information and warning information generated by the artificial intelligent early warning operating system 600 and is transmitted to a township artificial intelligent early warning system, and information data which is used for storing information data generated by the multidimensional man-machine-object cooperation interoperation system 620 of the risk factor management system 200, the system comprises a first-level artificial intelligence early warning system (1100), a second-level artificial intelligence early warning system (1200), a third-level artificial intelligence early warning system (700), an internet + distributed early warning kiosk (800), a fifth-level artificial intelligence early warning system (900), a fourth-level artificial intelligence early warning system (1000), a third-level artificial intelligence early warning system (1100), a second-level artificial intelligence early warning system (1200), and a fourth-level artificial intelligence early warning system (1300).

Artificial intelligence early warning operating system 600: an AI artificial intelligence early warning operation system developed based on Linux operation system architecture is used for managing and controlling computer running programs of computer hardware, software and data resources, is used for interfaces of all levels of artificial intelligence early warning systems communicated with an internet + distributed early warning pavilion, is used for interfaces of cloud computing, cloud storage, a cloud database, an artificial intelligence early warning system, an internet + distributed early warning pavilion and other software communication, is used for communication interfaces of a multi-dimensional man-machine-object cooperative interoperation system, mobile equipment and an intelligent television, is used for a man-machine interface to provide support for other application software and comprises a brain-like neural network system, a multi-dimensional man-machine-object cooperative interoperation system, a public safety intelligent monitoring early warning and prevention control system, an autonomous unmanned servo system, a world-integrated network information platform system, an intelligent physical connection and risk factor data acquisition system, A risk factor management system.

Artificial intelligence early warning server 700: the server is used for providing various high-performance computing services for a client in an early warning network system, and under the control of an artificial intelligence early warning operating system, the server provides services of centralized computing, information publishing and data management for a remote monitoring client through a network video server, an infrared night vision detection server, an infrared voice detection server, a video television wall server, a program controlled switch, an AI cloud computing server, an AI database server, a GPU cloud server, a Web server, a communication server, a display, a mixing matrix, a router and a modem which are connected with the server.

Internet + distributed early warning kiosk 800: namely an unattended early warning pavilion, which comprises a display, an artificial intelligent early warning server, a network video server, an infrared night vision detection server, an infrared voice detection server, an NVR network video recorder, a disk matrix, a video television wall server, a program controlled switch, a Web server, a communication server, a display, a high-definition network video decoder, a high-definition network digital matrix, a comprehensive management platform system, a streaming media server, an optical terminal, a router, a modem, a photovoltaic power generation module, a wireless charging receiver, a storage battery pack, a power distribution system, an air conditioning system, an access control system, a smoke sensing system, an infrared sensing system, an early warning factor control module, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module, an output early warning module and a wire mesh, and is used for placing hardware facilities of the artificial intelligent early warning system, for receiving risk factor information data collected by the monitoring and control point face recognition system 110, the voice recognition system 120, the dynamic recognition system 130, the biological feature recognition system 140, the infrared voice detection system 150, the infrared night vision detection system 160, the early warning unmanned aerial vehicle cruise system 170 and the intelligent internet of things security monitoring system 180, for outputting voice early warning information and image early warning information, for the five-level artificial intelligent early warning system 900 to look up the camera probe, the voice monitor, the infrared voice probe, the infrared night vision probe and the early warning unmanned aerial vehicle carpet type scanning information distributed at each control point, for receiving an operation instruction sent by the five-level artificial intelligent early warning system, for the early warning police kiosk to track the abnormal target location sent to the space-ground integrated information network platform system and collect risk factor data information, meanwhile, wide-area coverage, deep monitoring and global cooperation are carried out on the monitoring blind area of the distribution control point, safe and efficient real-time data transmission is carried out with the multi-dimensional man-machine object cooperative interoperation system 620, an auxiliary decision is provided for commanders, the early warning unmanned aerial vehicle can carry out real-time patrol on abnormal environments, and tracking and monitoring on abnormal behaviors are carried out.

Five-level artificial intelligence early warning system 900: the five-level artificial intelligence early warning system belongs to a village-tranquil level early warning system and comprises a television wall display, an artificial intelligence early warning server, a network video storage server, a video television wall server, an NVR network video recorder, a magnetic disk matrix, a video terminal, a high-definition network video decoder, a decoder service switch, a program-controlled switch, a network facility cabinet, a multipoint control unit MCU, a Web server, a communication server, a display, a high-definition network digital matrix, an optical terminal, a comprehensive management platform system, a streaming media server, a router, a modem, a UPS host, a UPS storage battery pack, an air conditioning system, an early warning factor control module, an equipment access gateway, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module and an output early warning module, and is used for monitoring risk source early warning information of the surrounding environment of the Internet + distributed early warning kiosk 800 in the, the early warning information comprises audio data and/or video data, is used for receiving the early warning information sent by the internet and the distributed early warning kiosk 800 and feeding back the early warning information to the four-stage artificial intelligent early warning system in time, is used for receiving an operation instruction sent by the four-stage artificial intelligent early warning system and replying an execution step, a process, a measure and an implementation result, is used for early warning the unmanned aerial vehicle to carry out real-time patrol on an abnormal environment and track and monitor abnormal behaviors, is used for carrying out data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization between the same IP address network segment and the mobile intelligent terminal equipment by the multi-dimensional man-machine-object cooperative interoperation system 620, is used for tracking the abnormal target location sent by the early warning kiosk to the world-integrated information network platform system, acquires risk factor data information and simultaneously covers a monitoring blind area of a distribution control, deep monitoring and global cooperation, safe and efficient real-time data transmission with the multi-dimensional man-machine object cooperative interoperation system 620, and auxiliary decision making is provided for commanders.

Four-stage artificial intelligence early warning system 1000: the four-level artificial intelligence early warning system belongs to a county-level early warning system and is used for monitoring risk source early warning information of the Internet + distributed early warning police kiosk 800, the five-level artificial intelligence early warning system 900 and the surrounding environment within the jurisdiction range, wherein the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by the Internet + distributed early warning kiosk 800 and the five-level artificial intelligence early warning system 900 and feeding back the early warning information to the three-level artificial intelligence early warning system in time, is used for receiving an operation instruction sent by the three-level artificial intelligence early warning system and replying an execution step, a process, a measure and an implementation result, is used for early warning an unmanned aerial vehicle to carry out real-time patrol on the abnormal environment and carry out tracking and monitoring on abnormal behaviors, and is used for carrying out data transmission, data transmission and data transmission between the same, The system comprises an analysis system, a display system, a UI switching system, an instruction execution system, a control system and an interoperation system, wherein the analysis system, the display system, the UI switching system, the instruction execution system, the control system and the interoperation system are synchronous, and the system is used for tracking abnormal target positioning sent to a heaven-earth integrated information network platform system by an early warning police kiosk, acquiring risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a deployment and control point, carrying out deep monitoring and global cooperation, carrying out safe and efficient real-time data transmission with.

Three-level artificial intelligence early warning system 1100: the three-level artificial intelligence early warning system belongs to a city level early warning system and is used for monitoring risk source early warning information of an internet + distributed early warning pavilion 800, a five-level artificial intelligence early warning system 900, a four-level artificial intelligence early warning system 1000 and the surrounding environment thereof in the jurisdiction range, wherein the early warning information comprises audio data and/or video data, is used for receiving the early warning information sent by the internet + distributed early warning pavilion 800, the five-level artificial intelligence early warning system 900 and the four-level artificial intelligence early warning system 1000 and feeding back the early warning information to the two-level artificial intelligence early warning system in time, is used for receiving an operating instruction sent by the two-level artificial intelligence early warning system and replying execution steps, processes, measures and implementation results, is used for carrying out real-time patrol on the abnormal environment by an early warning unmanned aerial vehicle, tracking and monitoring abnormal behaviors, and is used for carrying out data transmission between the same IP address network segment and mobile, The system comprises an analysis system, a display system, a UI switching system, an instruction execution system, a control system and an interoperation system, wherein the analysis system, the display system, the UI switching system, the instruction execution system, the control system and the interoperation system are synchronous, and the system is used for tracking abnormal target positioning sent to a heaven-earth integrated information network platform system by an early warning police kiosk, acquiring risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a deployment and control point, carrying out deep monitoring and global cooperation, carrying out safe and efficient real-time data transmission with.

Second-level artificial intelligence early warning system 1200: the second-stage artificial intelligence early warning system belongs to a provincial hall-level early warning system, is used for monitoring the risk source early warning information of an internet + distributed early warning kiosk 800, a fifth-stage artificial intelligence early warning system 900, a fourth-stage artificial intelligence early warning system 1000, a third-stage artificial intelligence early warning system 1100 and the surrounding environment within the jurisdiction, comprises audio data and/or video data, is used for receiving the early warning information sent by the internet + distributed early warning kiosk 800, the fifth-stage artificial intelligence early warning system 900, the fourth-stage artificial intelligence early warning system 1000 and the third-stage artificial intelligence early warning system 1100 and feeding back the early warning information to the first-stage artificial intelligence early warning system 1300 in time, is used for receiving an operating instruction sent by the first-stage artificial intelligence early warning system and replying an execution step, a process, a measure and an implementation result, is used for early warning an unmanned aerial vehicle to carry, the system is used for synchronizing data transmission, analysis, display, UI switching, instruction execution, control and interoperation of the multi-dimensional man-machine-object cooperative interoperation system 620 between the same IP address network segment and the mobile intelligent terminal device, is used for tracking abnormal target positioning sent by an early warning police kiosk to a world-integrated information network platform system, collecting risk factor data information, meanwhile, performing wide-area coverage on a monitoring blind area of a control point, performing deep monitoring and global cooperation, and performing safe and efficient real-time data transmission with the multi-dimensional man-machine-object cooperative interoperation system 620 to provide assistant decision-making for commanders.

First-level artificial intelligence early warning system 1300: the first-stage artificial intelligence early warning system belongs to a department-level early warning system and is used for monitoring risk source early warning information of an internet + distributed early warning kiosk 800, a fifth-stage artificial intelligence early warning system 900, a fourth-stage artificial intelligence early warning system 1000, a third-stage artificial intelligence early warning system 1100, a second-stage artificial intelligence early warning system 1200 and the surrounding environment thereof in the jurisdiction, wherein the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by the internet + distributed early warning kiosk 800, the fifth-stage artificial intelligence early warning system 900, the fourth-stage artificial intelligence early warning system 1000, the third-stage artificial intelligence early warning system 1100 and the second-stage artificial intelligence early warning system 1200 and feeding the early warning information back to the second-stage artificial intelligence early warning system 1200 in time, is used for sending operation instructions to the second-stage artificial intelligence early warning system and carrying out steps, processes and measures, the system is used for tracking and monitoring abnormal behaviors, carrying out data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization between a multi-dimensional man-machine-object cooperative interoperation system 620 and a mobile intelligent terminal device in the same IP address network segment, tracking abnormal target positioning sent by an early warning police kiosk to a heaven-earth integrated information network platform system, collecting risk factor data information, carrying out wide-area coverage on a monitoring blind area of a deployment point, carrying out deep monitoring and global cooperation, carrying out safe and efficient real-time data transmission with the multi-dimensional man-machine-object cooperative interoperation system 620, providing auxiliary decisions for commanders, carrying out vertical management and unified scheduling on an artificial intelligent early warning system, and stably mastering the global for maintaining frontier.

Preferably, the cloud storage system includes a face recognition blacklist.

Preferably, the risk factor collecting module collects image characteristic information and voice information, and the comparison result output module is a risk factor identifying module.

Preferably, the display module is a computer display screen, the resolution of the computer display screen is 1920 × 1080, the screen ratio is 16: 09, and the refresh rate is 144 Hz.

Preferably, the receiving optical transceiver and the transmitting optical transceiver are exchange signal enhancement modules.

Preferably, the camera is an AI camera probe.

Preferably, the cloud computing is designed based on an open-source Hadoop architecture.

Preferably, the cloud database: through the online data storage module, the online data storage module is designed based on an open-source Hadoop architecture.

Preferably, the cloud database is divided into an original voice information database, an original image characteristic information database, a real-time risk factor acquisition image information database, a real-time risk factor acquisition voice information database, a risk factor identification database, a risk factor reasoning database, a risk factor evaluation database, a risk factor response database, a risk factor management evaluation database, a real-time judgment basis database, a judgment rule database and an accident case database.

In an optimal mode, the artificial intelligence early warning operating system is an AI artificial intelligence early warning operating system developed on the basis of a Linux operating system architecture.

Preferably, the image feature information includes a collected face image, dynamic feature information, and biometric feature information.

Preferably, the voice information includes collected voice information.

Preferably, the original image feature information includes a face image blacklist, a dynamic feature information blacklist and a biological feature information blacklist stored in the storage module.

Preferably, the original voice message includes a voice message blacklist stored in the storage module.

Preferably, the daily activity evidence reasoning program is a risk factor reasoning module 230.

Preferably, the weekly activity evidence reasoning program is a risk factor reasoning module 230.

Preferably, the monthly activity evidence reasoning program is a risk factor reasoning module 230.

Preferably, the quarterly activity evidence reasoning program is a risk factor reasoning module 230.

Preferably, the annual activity evidence reasoning program is a risk factor reasoning module 230.

Preferably, the keyword recognition program is a risk factor inference module 230.

Preferably, the continuous speech recognition program is a risk factor inference module 230.

Preferably, the parser program is a risk factor inference module 230.

Preferably, the semantic information analysis program is a risk factor inference module 230.

Preferably, the risk factor inference module 230 is programmed by using a three-segment theory inference principle, wherein the three-segment theory inference principle includes a big premise, a small premise and a conclusion.

Preferably, the early warning value evaluation program is a risk factor evaluation module 240.

Preferably, the module 250 for generating early warning level alarm information, i.e. risk factor handling, is described.

Preferably, the generated early warning signal is fed back to the upper-level early warning system, i.e. the risk management evaluation module 260.

Preferably, the artificial intelligence early warning server 700 includes a high performance central processing unit CPU, an image processor GPU, a programmable gate array FPGA, a neural network processor NPU, and a heterogeneous/reconfigurable processor.

In a preferred manner, the internet + distributed pre-warning kiosk 800 is an unattended pre-warning kiosk.

In a preferable mode, the five-level artificial intelligence early warning system 900 belongs to a township early warning system.

In an optimal mode, the four-level artificial intelligence early warning system 1000 belongs to a county-level early warning system.

In a preferred mode, the three-level artificial intelligence early warning system 1100 belongs to a city-level early warning system.

In an optimal mode, the secondary artificial intelligence early warning system 1200 belongs to a provincial hall-level early warning system.

In an optimal mode, the primary artificial intelligence early warning system 1300 belongs to a department-level early warning system.

In a preferable mode, the early warning unmanned aerial vehicle cruise system 170 is controlled by a five-level artificial intelligence early warning system 900, namely a township early warning system control center.

Preferably, the brain-like neural network system 610 includes a deep neural network model for deep learning understanding of texts, images, audio, video, gene data and complex network data available for mass big data, and self-programming is completed by self-learning.

Preferably, the multidimensional man-machine-object cooperative interoperation system 620 updates and switches interfaces, forwards and displays messages, and analyzes and executes instructions to control and interoperate synchronously in the three-way interaction process of the man-machine-object, so as to complete intelligent sharing of multimedia resources.

Preferably, the public security intelligent monitoring, early warning, prevention and control system 630 includes a public security cloud computing solution, a crime information analysis solution, a crime prediction and prevention solution, and an intelligent video monitoring solution.

Preferably, the autonomous unmanned servo system 640 is an unmanned aerial vehicle background control center.

In a preferable mode, the space-ground integrated information network platform system 650 adopts a Beidou commercial system, namely a Beidou satellite navigation system.

Preferably, the photovoltaic power generation system 801 includes a monocrystalline silicon, a polycrystalline silicon and a thin film solar cell matrix.

Preferably, the UPS uninterruptible power supply 803, that is, the UPS, is used to provide stable and uninterrupted power supply for the manual intelligent early warning system host, the computer network system, the intelligent internet of things and the risk factor data acquisition system and subsystem when the public power grid is powered off.

Preferably, the intelligent access control system 812 is configured to set an authority for the access channel of the internet + distributed early warning kiosk 800.

Preferably, the video monitoring system 813 is used for providing security guarantee for monitoring the unattended internet and the distributed early warning kiosk 800 against human damage.

Preferably, the infrared sensing system 815 is used to prevent the internet and the distributed early warning kiosk 800 from being damaged by human beings, and to prevent extreme molecules from hitting, smashing, robbing and burning the kiosk to provide safety precaution.

Preferably, the early warning factor control module 816 is configured to control an alarm bell, an audible and visual alarm, an air conditioning system, an intelligent access control system, a video monitoring system, a smoke sensing system, and an infrared sensing system in the internet + distributed early warning kiosk 800.

In a preferable mode, the charged wire netting is installed in a range 5 meters away from the outside of the internet and the distributed early warning kiosk 800, and is used for preventing charged obstacles from being exceeded by a human body.

Drawings

Fig. 1 is a block diagram of an artificial intelligent early warning system architecture: 100. an intelligent Internet of things and risk factor data acquisition system; 200. a risk factor management system; 300. cloud computing; 400. cloud storage; 500. a cloud database; 600. an artificial intelligence early warning operating system; 700. an artificial intelligence early warning server; 800. the Internet and the distributed early warning pavilion; 900. a five-level artificial intelligence early warning system; 1000. a four-stage artificial intelligence early warning system; 1100. a three-level artificial intelligence early warning system; 1200. a secondary artificial intelligence early warning system; 1300. a primary artificial intelligence early warning system.

Fig. 2 is a schematic structural diagram of an intelligent internet of things and risk factor data acquisition system: 110. a face recognition system; 120. a speech recognition system; 130. a dynamic identification system; 140. a biometric identification system; 150. an infrared voice detection system; 160. an infrared night vision detection system; 170. an unmanned aerial vehicle cruise early warning system; 180. intelligent thing networking security protection monitored control system.

FIG. 3 is a schematic diagram of a risk factor management system: 210. a risk factor collection module; 220. a risk factor identification module; 230. a risk factor reasoning module; 240. a risk factor evaluation module; 250. a risk factor handling module; 260. and a risk management evaluation module.

Fig. 4 is a schematic view of risk source deployment and control of a risk factor data acquisition system: 1. the Internet and the distributed early warning pavilion; 2. a face recognition system camera probe; 3. dynamically recognizing a system camera probe; 4. a speech recognition system listener; 5. a camera probe of a biological characteristic recognition system; 6. an unmanned aerial vehicle cruise early warning system; 7. an infrared night vision detection system camera probe; 8. an infrared voice detection system probe; 9. an intelligent Internet of things security monitoring system sensor; 10. a video server; 11. a five-level artificial intelligent early warning system (village and peace level) monitors the staff of the center; 12. a network smart television wall; 13. the system comprises a mixing matrix, a video server and a comprehensive management system; 14. a communication data transmission system; 15. and a five-level artificial intelligent early warning system monitoring center.

FIG. 5 is a schematic diagram of an example of a module for collecting image feature information and voice information: the image characteristic information is compared with the original image characteristic information, analyzed and judged whether the image characteristic information is matched with the information of the image model base, if judged as 'Yes', early warning prompt information is generated, and then the signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if judged as 'Yes', the early warning value evaluation program generates early warning level warning information, namely a risk factor response module 250, the early warning level warning information is regenerated into early warning signals and is fed back to an upper early warning system, namely a risk management evaluation module 260, and the prompt information is transmitted to an artificial intelligent early warning operation system 600, if judged as 'No', the early warning value No evaluation result is transmitted to an image information database, namely a cloud database 500, and if compared, analyzed and compared with the original image characteristic information, and judged as 'No', the signal is transmitted to an activity reasoning evidence program, namely a risk factor reasoning module 230, the risk factor inference module 230: daily, weekly, monthly, quarterly and yearly activity sign reasoning program, if judging as "Yes", the daily activity sign reasoning program generates early warning prompt information, generates early warning prompt information and then transmits signals to an early warning value evaluation program, namely a risk factor evaluation module 240, if judging as "Yes", the early warning value evaluation program generates early warning level warning information, namely a risk factor response module 250, the early warning level warning information is regenerated into early warning signals and is fed back to a previous early warning system, namely a risk management evaluation module 260, and transmits the prompt information to an artificial intelligence early warning operation system 600, if judging as "No", the early warning value evaluation program transmits an early warning value No evaluation result to an image information database, namely a cloud database 500, and if judging as "No", the daily activity sign reasoning program transmits signals to a weekly activity sign reasoning program, namely a risk factor reasoning module 230, the collected image characteristic information is circularly inferred in activity signs of every day, every week, every month, every quarter and every year by the steps, the voice information and the original voice information are compared, analyzed and judged whether the voice information is matched with the information of the voice model base or not, if the voice information is judged to be 'Yes', early warning prompt information is generated, and then a signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module 250 is generated, the early warning level warning information is regenerated into an early warning signal and is fed back to an upper early warning system, namely a risk management evaluation module 260, and the prompt information is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program is judged to be 'No', then an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database 500, if the keyword recognition inference program judges to be "Yes", the keyword recognition inference program generates early warning prompt information, transmits the signal to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program judges to be "Yes", the early warning level alarm information, namely a risk factor response module 250, generates early warning signals, feeds the early warning signals back to an upper early warning system, namely a risk management evaluation module 260, and transmits the prompt information to an artificial intelligent early warning operating system 600, if the early warning value evaluation program judges to be "No", the early warning value No evaluation result is transmitted to a voice information database, namely a cloud database 500, if the keyword recognition inference program judges "No", a signal is transmitted to a continuous speech recognition inference program, i.e., a risk factor inference module 230, so as to perform loop inference of keyword recognition, continuous speech recognition, a grammar analyzer and a semantic information analysis inference program on the collected speech information.

FIG. 6 is a schematic structural diagram of an artificial intelligent early warning operating system: 610. a brain-like neural network system; 620. a multi-dimensional man-machine-object cooperative interoperation system; 630. a public safety intelligent monitoring, early warning, prevention and control system; 640. an autonomous unmanned servo system; 650. a space-ground integrated information network platform system.

FIG. 7 is a schematic structural diagram of a multi-dimensional man-machine-object cooperative interoperation system: the multi-dimensional man-machine-object cooperative interoperation system is used in the environment of cloud computing 300, cloud storage 400 and cloud database 500, and media servers authenticated by UPnP/DLNA are connected through bandwidth or WIFI network, data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization are carried out among different terminal devices, the method is used for accessing the mobile intelligent device and the intelligent television to the same local area network in the environment of cloud computing 300, cloud storage 400 and cloud database 500, setting IP addresses as the same network segment, autonomously discovering other devices and connecting the mobile intelligent device and the intelligent television through an interconnection mutual discovery module, and is used for a message control mechanism, the method comprises the steps of updating and switching an interface, forwarding and displaying messages, analyzing and executing control of instructions and synchronizing interoperation in the process of three-way interaction of human and animals, and completing intelligent sharing of multimedia resources.

Fig. 8 is a schematic structural diagram of a hardware system of the internet + distributed early warning kiosk: 801. a photovoltaic power generation system; 802. a battery pack; 803. a UPS (uninterrupted power supply); 804. a power distribution system; 805. a public power grid; 806. an expansion port; 807. a network adapter; 808. a fiber optic transceiver; 809. an optical fiber; 810. a communication module; 811. an air conditioning system; 812. an intelligent access control system; 813. a video monitoring system; 814. a smoke sensing system; 815. an infrared sensing system; 816. an early warning factor control module; 817. inputting an early warning factor module; 818. an intelligent early warning control center; 819. and an output early warning module.

Detailed Description

The technical scheme of the invention is clearly and completely described in the following description and the accompanying drawings.

The invention provides an artificial intelligence early warning system, which comprises an intelligent Internet of things and risk factor data acquisition system 100, and is used for acquiring face identification information, voice identification information, dynamic identification information, biological characteristic identification information, infrared voice detection information, infrared night vision detection information, early warning unmanned aerial vehicle cruise information and Internet of things intelligent security information of risk source factors in real time, judging risks and sending the acquired information to a risk factor management system, wherein the face identification information, the voice identification information, the dynamic identification information, the biological characteristic identification information, the infrared voice detection information, the infrared night vision detection information, the early warning unmanned aerial vehicle cruise information and.

The risk factor management system 200 is used for receiving the intelligent internet of things and risk factor data acquisition system, classifying the acquired face identification information, voice identification information, dynamic identification information, biological characteristic identification information, infrared voice detection information, infrared night vision detection information, early warning unmanned aerial vehicle cruise information and internet of things intelligent security information, and transmitting the classified image characteristic information and voice information to the risk factor identification module, wherein the image characteristic information comprises acquired face images, dynamic characteristic information and biological characteristic information; the voice information comprises collected voice information, the original image feature information comprises a face image blacklist, a dynamic feature information blacklist and a biological feature information blacklist which are stored in a storage unit, and the original voice information comprises a voice information blacklist which is stored in the storage unit.

The cloud computing 300 is designed based on an open-source Hadoop architecture, a user can develop a distributed program without knowing details of a distributed bottom layer, high-speed operation and storage are performed by using cluster advantages, and the cloud computing 300 is used for comparing and analyzing a risk factor identification module 220, a risk factor reasoning module 230 and a risk factor evaluation module 240 on a computing distributed computer with mass data, performing hierarchical reasoning and early warning value evaluation, and then returning a processing result to the user and performing cloud storage 400.

The cloud storage 400 is used for cluster application of a cloud computing system, integrates distributed system files through application software to cooperatively work, provides data storage and service access for users, the system is used for comparing the collected face image, dynamic characteristic information, biological characteristic information and voice information with the face image blacklist, the dynamic characteristic information blacklist, the biological characteristic information blacklist and the voice information blacklist in the storage module by arranging an online data storage module, wherein the face image blacklist, the dynamic characteristic information blacklist, the biological characteristic information blacklist and the voice information blacklist are stored in the storage module, if the similarity reaches a preset early warning value, the early warning system generates early warning prompt information for risk factor reasoning and evaluation in time, generates early warning level warning information and feeds the early warning level warning information back to the previous early warning system for risk management and evaluation.

The cloud database 500 is divided into an original voice information database, an original image characteristic information database, an original information database, a real-time risk factor acquisition image information database, a real-time risk factor acquisition voice information database, a risk factor identification database, a risk factor reasoning database, a risk factor evaluation database, a risk factor response database, a risk factor management evaluation database, a real-time judgment basis database, a judgment rule database and an accident case database, the cloud database 500 is used for storing risk factor information data acquired by the intelligent things-tie and risk factor data acquisition system 100, the system is used for storing voice information contrastive analysis, keyword recognition, continuous voice recognition, a grammar analyzer, semantic information analysis and image characteristic information contrastive analysis, daily activity sign reasoning, weekly activity sign reasoning, monthly activity sign reasoning, quarterly activity sign reasoning, annual activity sign reasoning, early warning prompt information generation, early warning value evaluation, early warning level warning information generation, early warning signal generation feedback to an upper-level early warning system, information data which is updated in real time according to data transmitted by a cloud computing 300 through a cloud storage 400, information data which is used for storing prompt information, grading information, early warning information and warning information generated by the artificial intelligent early warning operating system 600 and transmitted to a township artificial intelligent early warning system, and information data generated by the multi-dimensional man-machine-object cooperation interoperation system 620 of the risk factor management system 200, the system comprises a first-level artificial intelligence early warning system (1100), a second-level artificial intelligence early warning system (1200), a third-level artificial intelligence early warning system (700), an internet + distributed early warning kiosk (800), a fifth-level artificial intelligence early warning system (900), a fourth-level artificial intelligence early warning system (1000), a third-level artificial intelligence early warning system (1100), a second-level artificial intelligence early warning system (1200), and a fourth-level artificial intelligence early warning system (1300).

An artificial intelligence early warning operation system 600, an AI artificial intelligence early warning operation system developed based on Linux operation system architecture, the system is used for managing and controlling computer running programs of computer hardware, software and data resources, used for interfaces of all levels of artificial intelligence early warning systems communicating with internet and distributed early warning kiosks, used for interfaces of cloud computing, cloud storage, cloud databases and artificial intelligence early warning systems, internet and distributed early warning kiosks and other software communication, used for communication interfaces of a multidimensional man-machine-object cooperative interoperation system, mobile equipment and an intelligent television, and used for a man-machine interface to provide support for other application software, and the system comprises a brain-like neural network architecture system, a multidimensional man-machine-object cooperative interoperation system, a public safety intelligent monitoring early warning and prevention and control system, an autonomous unmanned servo system, a world-integrated network information platform system, The system comprises an intelligent Internet of things and risk factor data acquisition system and a risk factor management system.

The artificial intelligence early warning server 700 is used for providing various high-performance computing services for clients in the early warning network system, and under the control of an artificial intelligence early warning operating system, the server provides services of centralized computing, information publishing and data management for remote monitoring clients through a network video server, an infrared night vision detection server, an infrared voice detection server, a video television wall server, a program control switch, an AI cloud computing server, an AI database server, a GPU cloud server, a Web server, a communication server, a display, a hybrid matrix, a router and a modem which are connected with the server.

The internet + distributed early warning kiosk 800, namely an unattended early warning kiosk, is used for placing hardware facilities of an artificial intelligence early warning system, is used for receiving risk factor information data collected by the monitoring and control point face recognition system 110, the voice recognition system 120, the dynamic recognition system 130, the biological feature recognition system 140, the infrared voice detection system 150, the infrared night vision detection system 160, the early warning unmanned aerial vehicle cruise system 170 and the intelligent internet of things security monitoring system 180, is used for outputting voice early warning alarm information and image early warning alarm information, is used for the five-stage artificial intelligence early warning system 900 to look up camera probes, voice monitors, infrared voice probes, infrared night vision probes and early warning unmanned aerial vehicle blanket type scanning information distributed at various control points, is used for receiving operation instructions sent by the five-stage artificial intelligence early warning system, and is used for tracking abnormal target positioning sent by the early warning kiosk to a world integration information network platform system, risk factor data information is collected, meanwhile, wide-area coverage is conducted on a monitoring blind area of a distribution control point, deep monitoring and global cooperation are conducted, safe and efficient real-time data transmission is conducted with the multi-dimensional man-machine object cooperative interoperation system 620, an auxiliary decision is provided for commanders, the early warning unmanned aerial vehicle can conduct real-time patrol on abnormal environments, and tracking and monitoring on abnormal behaviors are conducted.

A five-level artificial intelligence early warning system 900, which belongs to a township early warning system and is used for monitoring risk source early warning information of the surrounding environment of the internet + distributed early warning kiosk 800 within the jurisdiction, wherein the early warning information comprises audio data and/or video data, is used for receiving the early warning information sent by the internet + distributed early warning kiosk 800 and feeding back the early warning information to the four-level artificial intelligence early warning system in time, is used for receiving an operation instruction sent by the four-level artificial intelligence early warning system and replying execution steps, processes, measures and execution results, is used for early warning the unmanned aerial vehicle to patrol the abnormal environment in real time, tracking and monitoring abnormal behaviors, and is used for the man-machine-object cooperation multi-dimensional interoperation system 620 to perform data transmission, analysis, display, UI switching, instruction execution network segment, and, The control and the interoperation are synchronous, and the system is used for tracking abnormal target positioning sent by the early warning police kiosk to the heaven-earth integrated information network platform system, acquiring risk factor data information, performing wide-area coverage on a monitoring blind area of a control point, performing deep monitoring and global cooperation, transmitting safe and efficient real-time data with the multi-dimensional man-machine object cooperative interoperation system 620, and providing assistant decision-making for commanders.

The four-level artificial intelligence early warning system 1000 belongs to a county-level early warning system and is used for monitoring risk source early warning information of the internet + distributed early warning kiosk 800, the five-level artificial intelligence early warning system 900 and the surrounding environment within the jurisdiction range, wherein the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by the internet + distributed early warning kiosk 800 and the five-level artificial intelligence early warning system 900 and feeding back the early warning information to the three-level artificial intelligence early warning system in time, is used for receiving an operation instruction sent by the three-level artificial intelligence early warning system and replying an execution step, process, measure and implementation result, is used for early warning an unmanned aerial vehicle to carry out real-time patrol on abnormal environment and trace and monitor abnormal behaviors, and is used for carrying out data transmission between the same IP address network segment and the mobile intelligent terminal equipment by the multi-dimensional man-machine cooperation operation, The system comprises an analysis system, a display system, a UI switching system, an instruction execution system, a control system and an interoperation system, wherein the analysis system, the display system, the UI switching system, the instruction execution system, the control system and the interoperation system are synchronous, and the system is used for tracking abnormal target positioning sent to a heaven-earth integrated information network platform system by an early warning police kiosk, acquiring risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a deployment and control point, carrying out deep monitoring and global cooperation, carrying out safe and efficient real-time data transmission with.

A third-level artificial intelligence early warning system 1100, which belongs to a city-level early warning system and is used for monitoring risk source early warning information of the internet + distributed early warning kiosk 800, the fifth-level artificial intelligence early warning system 900, the fourth-level artificial intelligence early warning system 1000 and the surrounding environment thereof in the jurisdiction range, wherein the early warning information comprises audio data and/or video data, is used for receiving the early warning information sent by the internet + distributed early warning kiosk 800, the fifth-level artificial intelligence early warning system 900 and the fourth-level artificial intelligence early warning system 1000 and feeding back the early warning information to the second-level artificial intelligence early warning system in time, is used for receiving an operation instruction sent by the second-level artificial intelligence early warning system and replying execution steps, processes, measures and implementation results, is used for performing real-time patrol on abnormal environment by an early warning unmanned aerial vehicle and tracking and monitoring abnormal behaviors, and is used for performing tracking and monitoring on the abnormal behaviors between the same IP address network segment and a The system comprises a data transmission system, a data analysis system, a display system, a UI switching system, an instruction execution system, a control system and an interoperation system, wherein the data transmission system, the analysis system, the display system, the UI switching system, the instruction execution system, the control system and the interoperation system are synchronous, and are used for tracking abnormal target positioning sent by an early warning police kiosk to a space-ground integrated information network platform system, acquiring risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a distribution control point, carrying out deep monitoring and global cooperation, carrying out safe and efficient real-.

The face recognition system 110 is used for extracting the face feature points in the process of processing the image video by the computer, and performing analysis, identity recognition, modeling and retrieval by utilizing the principle of biometrical to find out whether the specified person exists.

The speech recognition system 120 is used for implementing operations of speech recognition, voice control, man-machine conversation and dynamic keyword editing, and is used for recognizing the voice of a specific person and recognizing the voices of multiple persons.

The dynamic recognition system 130 is used for carrying out dynamic portrait detection and image extraction on pedestrians passing through the camera, the pedestrians walk through a recognition area in a natural form, the camera can carry out snapshot and collection of information and send corresponding commands to carry out dynamic face recognition, gait recognition is used in the occasion of video intelligent monitoring, the system carries out an identity recognition process on the walking posture of the pedestrians, the gait recognition system comprises a monitoring camera, a computer, a gait video/image sequence processing system and recognition software, the monitoring camera collects the gait of the pedestrians, a video sequence of the gait is obtained through detection and tracking, the gait features of the pedestrians are extracted through preprocessing analysis, the gait motion in the image sequence is subjected to the key steps of motion detection, motion segmentation and feature extraction and gait recognition, and then the gait features are stored in a database in a gait mode through calculation processing, and then, comparing and identifying the newly acquired gait features with the gait features of the gait database, performing pre-alarming or alarming if the newly acquired gait features are matched, and continuing acquiring the gait if the monitoring camera is not matched with the gait features.

The biological characteristic recognition system 140 is used for identifying the personal identity of the inherent physiological characteristics and behavior characteristics of a human body, firstly, the biological characteristics are sampled, the unique characteristics of the biological characteristics are extracted and converted into digital codes, the codes are further combined into a characteristic template, and when people interact with the recognition system to carry out identity authentication, the recognition system acquires the characteristics of the biological characteristics and compares the characteristics with the characteristic template in a database to determine whether the biological characteristics are matched, so that the people are determined to be accepted or rejected.

An infrared voice detection system 150 is an infrared laser device for remotely monitoring voices of a suspicious target room, and is characterized in that firstly, a laser beam is used for irradiating the glass of the target room at a certain distance outside the target room, the reflected laser beam is modulated by a glass vibration signal caused by sound pressure, a photoelectric sensor is used for receiving the reflected laser beam and converting the reflected laser beam into an electric signal, a circuit is used for filtering and amplifying the obtained signal, and meanwhile, audio processing software Adobe audio filtering is used for being matched with hardware filtering to optimize and restore the signal again, so that the monitoring of the target is realized.

The infrared night vision detection system 160 is used for actively projecting infrared light onto an object by adopting an infrared emitting device in a dark environment without visible light or dim light, and the infrared light enters a lens for imaging after being reflected by the object to monitor the night action trace of suspicious target personnel.

Early warning unmanned aerial vehicle system 170 that cruises for the collection of high resolution image when making up satellite remote sensing and often sheltering from the shortcoming that can not obtain the image because of the cloud cover, has solved traditional satellite remote sensing and has revisited the cycle length again, emergent untimely problem, and unmanned aerial vehicle is equipped with high definition digital video camera and GPS positioning system, can fix a position independently and cruise, and the image is shot in real-time conveying, and the control personnel can watch and control in step on the computer.

An intelligent Internet of things security monitoring system 180 is characterized in that the intelligent security Internet of things comprises sensor nodes, a gateway, router nodes and a security executing mechanism based on a wireless sensor network, the sensor nodes are used for collecting Internet of things information and transmitting various security sensor data, generating alarm signals after processing the data and informing users of timely processing emergencies, and controlling the executing mechanism to act simultaneously, the router nodes are used for managing the network and are also used as nodes of sensors, distributing network addresses for sub-nodes of the router nodes and helping other nodes to route the data to be sent to destination nodes, the network range is greatly expanded, the security executing mechanism is used for the bottommost layer of the intelligent security Internet of things, is controlled by the sensor nodes and is mainly used for executing and solving various security problems, and the transmission network is used for transmitting a broadband network and a security Internet of things platform, And the transmission of the Internet of things of the telecommunication gateway is used for realizing the transmission and calculation of data.

The risk factor collecting module 210 is configured to find out a predicted uncertain factor, analyze an environmental condition and a sensitivity degree to an early warning scheme, evaluate related data, calculate an effect under various risk conditions, and make a correct judgment.

The risk factor identification module 220 is used for identifying potential risks and behavior characteristics thereof, identifying main sources of the risks and predicting possible consequences caused by the risks.

And the risk factor reasoning module 230 is used for finding out the main risk faced by the security subject from the complicated environment.

The risk factor evaluation module 240 is used for performing a quantitative evaluation process on the possibility that a safety event possibly occurs to influence and lose various aspects of life, life and property of people before the risk event occurs, and evaluating the possible degree of influence or loss caused by a certain event or thing.

A risk factor handling module 250 for avoiding future security events, for accepting risks to maintain an existing risk level for static viewing, for reducing risks, and for reducing risks to an acceptable level using policies or measures.

And the risk management evaluation module 260 is used for analyzing, checking, correcting and evaluating the applicability and effectiveness of the risk management technology.

A risk factor data acquisition system risk source deployment and control schematic diagram, which comprises 1, internet and a distributed early warning kiosk; 2. a face recognition system camera probe; 3. dynamically recognizing a system camera probe; 4. a speech recognition system listener; 5. a camera probe of a biological characteristic recognition system; 6. an unmanned aerial vehicle cruise early warning system; 7. an infrared night vision detection system camera probe; 8. an infrared voice detection system probe; 9. an intelligent Internet of things security monitoring system sensor; 10. a video server; 11. a fifth-level artificial intelligent early warning system monitors the staff of the center; 12. a network smart television wall; 13. the system comprises a mixing matrix, a video server and a comprehensive management system; 14. a communication data transmission system; 15. and a five-level artificial intelligent early warning system monitoring center.

An example schematic diagram of a module for collecting image characteristic information and voice information, wherein the image characteristic information and original image characteristic information are compared, analyzed and judged whether the image characteristic information is matched with information of an image model base or not, if the image characteristic information is judged to be 'Yes', early warning prompt information is generated, the early warning prompt information is generated and then transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module 250 is generated, the early warning level warning information is regenerated into early warning signals and is fed back to a previous early warning system, namely a risk management evaluation module 260, the prompt information is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program is judged to be 'No', an early warning value No evaluation result is transmitted to an image information database, namely a cloud database 500, and compared, analyzed and compared with the original image characteristic information, and if the image characteristic Namely a risk factor reasoning module 230, the risk factor reasoning module is used for a daily, weekly, monthly, quarterly and annual activity sign reasoning program, if the daily activity sign reasoning program judges as "Yes", early warning prompting information is generated, and then a signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program judges as "Yes", early warning level warning information, namely a risk factor handling module 250, is generated, the early warning level warning information is regenerated into an early warning signal which is fed back to an upper-level early warning system, namely a risk management evaluation module 260, and the prompting information is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program judges as "No", an early warning value No evaluation result is transmitted to an image information database, namely a cloud database 500, and if the daily activity sign reasoning program judges as "No", the signal is transmitted to the weekly activity sign reasoning module 230, the collected image characteristic information is circularly inferred in activity signs of every day, every week, every month, every quarter and every year, the voice information and the original voice information are contrastingly analyzed to judge whether the voice information is matched with the information of the voice model base or not, if the voice information is judged to be 'Yes', early warning prompt information is generated, and then signals are transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module 250 is generated, the early warning level warning information is regenerated into early warning signals which are fed back to an upper early warning system, namely a risk management evaluation module 260, and the prompt information is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program is judged to be 'No', then an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database 500, if the keyword recognition inference program judges "Yes", an early warning prompt message is generated, and then the signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module 240, if the early warning value evaluation program judges "Yes", an early warning level warning message, namely a risk factor response module 250, the early warning level warning message is regenerated into an early warning signal and is fed back to an upper early warning system, namely a risk management evaluation module 260, and the prompt message is transmitted to an artificial intelligent early warning operation system 600, if the early warning value evaluation program judges "No", an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database 500, if the keyword recognition inference program judges "No", a signal is transmitted to a continuous speech recognition inference program, i.e., a risk factor inference module 230, so as to perform loop inference of keyword recognition, continuous speech recognition, a grammar analyzer and a semantic information analysis inference program on the collected speech information.

The brain-like neural network system 610 is used for deeply learning understanding of texts, images, audios, videos, gene data and complex network data which are available for massive large data, self-programming is completed through self-learning, the cognitive neural network model is used for the neural network to convert visual, sensory, perceptual, memory, thinking, imagination, reasoning and language information into internal mental activities through information processing and processing, the process of dominant behavior output is carried out, the perception understanding decision neural network model is used for the application of brain-like intelligent behaviors to environment perception, data understanding and reasoning decision making capacity, and the brain-like neural network system can adapt to environment, summarize rules, complete operation, recognition or process control.

The multi-dimensional man-machine-object cooperative interoperation system 620 is used in the environments of the cloud computing 300, the cloud storage 400 and the cloud database 500, and is connected with a media server authenticated by UPnP/DLNA through a bandwidth or WIFI network, data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization are carried out among different terminal devices, the method is used for accessing the mobile intelligent device and the intelligent television to the same local area network in the environment of cloud computing 300, cloud storage 400 and cloud database 500, setting IP addresses as the same network segment, autonomously discovering other devices and connecting the mobile intelligent device and the intelligent television through an interconnection mutual discovery module, and is used for a message control mechanism, the method comprises the steps of updating and switching an interface, forwarding and displaying messages, analyzing and executing control of instructions and synchronizing interoperation in the process of three-way interaction of human and animals, and completing intelligent sharing of multimedia resources.

A public safety intelligent monitoring, early warning, prevention and control system 630 is used for acquiring and processing a cloud platform through cloud computing information with ultra-strong high bandwidth and high concurrent information processing performance, realizing seamless, all-weather and multi-way safety information and information acquisition, wherein the acquired objects are aerospace acquired data, satellite remote sensing data, electromagnetic acquired data, video monitoring acquired data and night infrared detection acquired data based on an Internet of things platform, realizing real-time cleaning and processing integration of mass safety information through ultra-high performance mass information fusion and the cloud computing processing platform, realizing integration, high efficiency and ordered information access, management and sharing of mass safety information and GIS data high-value information data through an ultra-large and ultra-high capacity information storage and management cloud computing platform, and researching and judging the cloud computing platform through ultra-high performance information, data analysis and prediction, and the intelligent aid decision platform realizes intelligent visual display of safety information, aid analysis of leadership decision and implementation of leadership decision.

Autonomous unmanned servo system 640 is unmanned aerial vehicle backstage control center, unmanned aerial vehicle is used for carrying out image acquisition to camera control blind area, data transmission, patrol in real time to unusual environment, track monitoring unusual action, carry out identity tracking to suspicious personnel, the place is tracked, the action is tracked, the control station is that unmanned aerial vehicle is on ground, the flight control center of machine-borne or ship-borne, realize man-machine interaction, data transmission, watch in step and control, six rotor unmanned aerial vehicle essential function divide into hover in the air, locate function, predetermine the flight of airline, the remote control flight, the flight of autonomy, remote data transmission and flight state monitoring, visual identification and unmanned aerial vehicle servo function, the flight of ground station, system fault early warning function.

A space-ground integrated information network platform system 650 is a Beidou satellite navigation system which is a Beidou commercial system and is used for public security services such as anti-terrorism, stability maintenance, security, and security protection, and comprises applications such as public security vehicle command and dispatch, civil police field law enforcement, emergency event information transmission, public security time service, vehicle autonomous navigation, vehicle tracking monitoring, vehicle intelligent information system, and vehicle networking applications, wherein the emergency event information transmission uses a Beidou specific short message function, and a Beidou/GNSS high-precision receiver is used for providing 1-2 meter, decimeter and centimeter-level real-time high-precision navigation and positioning services in a service area through a ground reference station network by using broadcasting means such as satellites, mobile communication and digital broadcasting.

And the photovoltaic power generation system 801 is used for supplying a reserve power supply for the internet and the distributed early warning police kiosk 800 when the public power grid is suddenly powered off.

And the storage battery pack 802 is used for storing electric energy generated when the solar cell matrix is illuminated and supplying power to a load at any time.

The UPS 803 is used for providing stable and uninterrupted power supply for a host computer of an artificial intelligent early warning system, a computer network system, an intelligent Internet of things, a risk factor data acquisition system and a subsystem, when the mains supply input is normal, the UPS supplies the mains supply to a load after stabilizing the voltage, the UPS is an alternating current type voltage stabilizer and also charges a battery in the machine, and when the mains supply is interrupted (accident power failure), the UPS immediately converts the direct current electric energy of the battery into 220V alternating current through an inverter to supply power to the load continuously, so that the load keeps normal work and protects load software and hardware from being damaged.

And the power distribution system 804 is used for providing power for the internet and the distributed early warning kiosk 800.

A public power grid 805 for an electric power network system that transforms voltage and distributes power directly to end users for the internet + distributed early warning kiosks 800.

The expansion port 806 is used for the interface of the motherboard to connect various external devices, and through these expansion ports, almost all external devices such as a printer, an external optical floppy drive, a scanner, a flash disk, a mouse, a mobile phone, a digital camera, a mobile hard disk, a USB network card, a camera, a keyboard, a mouse, an ADSL Modem, and a Cable Modem can be connected.

A network adapter 807 for connecting computers and transmission media in a lan, and also for enabling physical connection and electrical signal matching between lan transmission media, and also relates to the transmission and reception of frames, the encapsulation and decapsulation of frames, media access control, encoding and decoding of data, and the buffering of data.

And the optical fiber transceiver 808 is used for conversion between optical signals and electric signals and transmission of high-definition video images in a network communication system.

And an optical fiber 809 for communication transmission.

And the communication module 810 is used for transmitting data through a GPRS and short message dual channel and supporting multi-center data communication.

The air conditioning system 811 is used for processing the temperature, humidity, cleanliness and airflow speed of indoor air for the internet + distributed early warning kiosk 800 so as to meet the normal operation of the early warning system in the kiosk.

And the intelligent access control system 812 is used for setting the authority for the access channel of the internet + distributed early warning police kiosk 800.

The video monitoring system 813 is used for preventing artificial damage and providing safety guarantee for monitoring the unattended internet and the distributed early warning police kiosk 800.

And the smoke sensing system 814 is used for preventing the monitoring host in the internet + distributed early warning kiosk 800 from being short-circuited or a fire is caused by spontaneous combustion of a line.

And the infrared induction system 815 is used for preventing the internet and the distributed early warning police box 800 from being damaged artificially and preventing extreme molecules from beating, smashing, robbing and burning the police box to provide safety precaution measures.

And the early warning factor control module 816 is used for controlling an alarm bell, an audible and visual alarm, an air conditioning system, an intelligent access control system, a video monitoring system, a smoke sensing system and an infrared sensing system in the internet + distributed early warning kiosk 800, wherein an output relay provides two normally open and two normally closed switching electric shocks, the contact capacities of the two normally open and two normally closed switching electric shocks are AC250V/3A and DC24V/5A, and the module receives a passive normally open feedback signal.

And the input early warning factor module 817 is used for outputting a signal to the control host, or providing a switching value signal to enable the controlled equipment to operate, and meanwhile, receiving a feedback signal of the equipment so as to report to the host.

And an intelligent early warning control center 818, configured to monitor the outdoor security status of the internet + distributed early warning kiosk 800 and the front-end control device of the indoor artificial intelligent early warning system.

The output early warning module 819 is used for switching between normal monitoring and early warning signals of a bus-system security early warning system, a microprocessor is embedded in the output early warning module, the microprocessor is communicated with the early warning controller, power-down detection of a power bus, fault detection of an input/output line, output control, logic state judgment of an input signal and control of a state indicator lamp are achieved, after the module receives a starting command of the early warning controller, a relay is attracted, the normal monitoring state is switched to the early warning state on site, the indicator lamp is lightened, and meanwhile, answer signal information is transmitted to the early warning controller to indicate that the switching is successful.

Claims

1. An artificial intelligence early warning system which characterized in that: the method comprises the following steps:

intelligent Internet of things and risk factor data acquisition system (100): the system comprises a face recognition system (110), a voice recognition system (120), a dynamic recognition system (130), a biological feature recognition system (140), an infrared voice detection system (150), an infrared night vision detection system (160), an early warning unmanned aerial vehicle cruise system (170) and an intelligent Internet of things security monitoring system (180), and is used for collecting information and sending the collected information to a risk factor management system (200);

risk factor management system (200): the risk factor management system comprises a risk factor collection module (210), a risk factor identification module (220), a risk factor reasoning module (230), a risk factor evaluation module (240), a risk factor correspondence module (250) and a risk management evaluation module (260); the risk factor management system (200) is used for receiving the intelligent Internet of things and risk factor data acquisition system (100), classifying the acquired face identification information, voice identification information, dynamic identification information, biological feature identification information, infrared voice detection information, infrared night vision detection information, early warning unmanned aerial vehicle cruise information and Internet of things intelligent security information, and transmitting the classified image feature information and voice information to the risk factor identification module (220), wherein the image feature information comprises acquired face images, dynamic feature information and biological feature information; the voice information comprises collected voice information, the original image characteristic information comprises a face image blacklist, a dynamic characteristic information blacklist and a biological characteristic information blacklist which are stored in a storage module, and the original voice information comprises a voice information blacklist which is stored in the storage module;

the image characteristic information and the original image characteristic information are compared, analyzed and judged to judge whether the image characteristic information is matched with information of an image model base or not, if the image characteristic information is judged to be 'Yes', early warning prompt information is generated, the early warning prompt information is generated and then is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module (250) is generated, the early warning level warning information is regenerated into early warning signals and is fed back to a previous early warning system, namely a risk management evaluation module (260), the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program is judged to be 'No', an early warning value No evaluation result is transmitted to an image information database, namely a cloud database (500), and if the image characteristic information is compared, analyzed and judged to be 'No', the signals are transmitted to an daily activity evidence reasoning ) (ii) a

If the daily activity evidence reasoning program judges that the daily activity evidence reasoning program is 'Yes', early warning prompt information is generated, and then the early warning prompt information is generated and then a signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges that the daily activity evidence reasoning program is 'Yes', early warning level warning information, namely a risk factor corresponding module (250), is generated, and then early warning level warning information is generated and fed back to an upper-level early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges that the daily activity evidence reasoning program is 'No', then an early warning value No evaluation result is transmitted to an image information database, namely a cloud database (500), and if the daily activity evidence reasoning program judges that the daily activity evidence reasoning program is;

if the weekly activity sign reasoning program judges as 'Yes', early warning prompt information is generated, and then the signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges as 'Yes', early warning level warning information, namely a risk factor corresponding module (250), is generated, the early warning level warning information is then generated into an early warning signal and is fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges as 'No', an early warning value No evaluation result is transmitted to an image information database, namely a cloud database (500), and if the weekly activity sign reasoning program judges as 'No', the signal is transmitted to a monthly activity sign reasoning program, namely a risk factor reasoning module (230);

if the monthly activity evidence reasoning program judges that the monthly activity evidence reasoning program is 'Yes', early warning prompt information is generated, and then a signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges that the monthly activity evidence reasoning program is 'Yes', an early warning level warning message, namely a risk factor corresponding module (250) is generated, the early warning level warning message is generated into an early warning signal and is fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges that the monthly activity evidence reasoning program is 'No', an early warning value No evaluation result is transmitted to an image information database, namely a cloud database (600), and if the monthly activity evidence reasoning program judges that the monthly activity evidence reasoning program is 'No', the signal is transmitted to;

if the quarterly activity evidence reasoning program judges as 'Yes', early warning prompt information is generated, and then signals are transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges as 'Yes', early warning level warning information, namely a risk factor response module (250), is generated, and then the early warning level warning information is generated into early warning signals to be fed back to a previous early warning system, namely a risk management evaluation module (260), and transmits the prompt information to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges as 'No', then an early warning value No evaluation result is transmitted to an image information database, namely a cloud database (500), and if the quarterly activity evidence reasoning program judges as 'No', then signals are transmitted to an annual activity reasoning evidence program, namely a risk factor reasoning module (230);

if the annual activity evidence reasoning program judges that the result is 'Yes', early warning prompt information is generated, the early warning prompt information is generated and then transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges that the result is 'Yes', early warning level warning information, namely a risk factor corresponding module (250), the early warning level warning information is generated and then fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges that the result is 'No', an early warning value No evaluation result is transmitted to an image information database, namely a cloud database (500), and if the annual activity evidence reasoning program judges that the result is 'No', image characteristic information acquired is transmitted to the image information database, namely the cloud database (500);

the voice information and the original voice information are compared, analyzed and judged to judge whether the voice information is matched with the information of the voice model base or not, if the voice information is judged to be 'Yes', early warning prompt information is generated, and then the signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program is judged to be 'Yes', early warning level warning information, namely a risk factor response module (250) is generated, the early warning level warning information is regenerated into early warning signals and is fed back to an upper early warning system, namely a risk management evaluation module (260), the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program is judged to be 'No', a No warning value No evaluation result is transmitted to a voice information database, namely a cloud database (500), if the voice information is compared with the original voice information, the voice information is analyzed, and if the voice information is judged to be No, a signal is transmitted to a keyword recognition program, namely a risk factor reasoning module (230);

if the keyword recognition program judges as 'Yes', early warning prompt information is generated, and then a signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges as 'Yes', early warning level warning information, namely a risk factor corresponding module (250), is generated, the early warning level warning information is regenerated into an early warning signal and is fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges as 'No', an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database (500), and if the keyword recognition program judges as 'No', the signal is transmitted to a continuous voice recognition program, namely a risk factor reasoning module (230);

if the continuous voice recognition program judges Yes, early warning prompt information is generated, and then signals are transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges Yes, early warning level warning information, namely a risk factor corresponding module (250) is generated, the early warning level warning information is regenerated into early warning signals and is fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges No, then an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database (500), and if the continuous voice recognition program judges No, then the signals are transmitted to a grammar analyzer program, namely a risk factor reasoning module (230);

if the grammar analyzer program judges to be 'Yes', early warning prompt information is generated, and then a signal is transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges to be 'Yes', early warning level warning information, namely a risk factor corresponding module (250), is generated, the early warning level warning information is regenerated into an early warning signal and is fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges to be 'No', an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database (500), and if the grammar analyzer program judges to be 'No', the signal is transmitted to a semantic information analysis program, namely a risk factor reasoning module (230);

if the semantic information analysis program judges that the semantic information analysis program is 'Yes', early warning prompt information is generated, and then the early warning prompt information is generated and then transmitted to an early warning value evaluation program, namely a risk factor evaluation module (240), if the early warning value evaluation program judges that the semantic information analysis program is 'Yes', early warning level warning information, namely a risk factor corresponding module (250), is generated, the early warning level warning information is regenerated into an early warning signal and is fed back to a previous early warning system, namely a risk management evaluation module (260), and the prompt information is transmitted to an artificial intelligent early warning operation system (600), if the early warning value evaluation program judges that the semantic information analysis program is 'No', an early warning value No evaluation result is transmitted to a voice information database, namely a cloud database (500), and if the semantic information analysis program judges that the semantic information analysis program is 'No', the acquired voice information;

cloud computing (300): the method is designed based on an open-source Hadoop architecture, a user can develop a distributed program under the condition that distributed bottom-layer details are not known, high-speed operation and storage are carried out by utilizing cluster advantages, a cloud computing (300) comprises infrastructure as a service, a platform as a service and software as a service, and is used for a risk factor identification module (220), a risk factor reasoning module (230) and a risk factor evaluation module (240) on a computing distributed computer, a huge computing processing program is automatically divided into numerous small subprograms through a network, the subprograms are delivered to a huge system consisting of a plurality of servers to be compared and analyzed with massive data information through searching, hierarchical reasoning and early warning value evaluation are carried out, and then a processing result is returned to the user and cloud storage (400) is carried out; cloud storage (400): the system comprises a cloud storage server, cloud storage application software and distributed storage equipment, is used for cluster application of a cloud computing (300) system, integrates distributed system files through the application software to cooperatively work to provide data storage and service access for users, is used for setting an online data storage module, stores a face image blacklist, a dynamic characteristic information blacklist, a biological characteristic information blacklist and a voice information blacklist in the storage module, compares the collected face image, dynamic characteristic information, biological characteristic information and voice information with the face image blacklist, the dynamic characteristic information blacklist, the biological characteristic information blacklist and the voice information blacklist in the storage module, and if the similarity reaches a preset early warning value, an early warning system timely generates early warning prompt information to carry out risk factor reasoning, Evaluating, generating early warning level alarm information, and feeding back to an upper-level early warning system for risk management evaluation;

cloud database (500): the cloud database (500) is divided into an original voice information database, an original image characteristic information database, a real-time risk factor acquisition image information database, a real-time risk factor acquisition voice information database, a risk factor identification database, a risk factor reasoning database, a risk factor evaluation database, a risk factor response database, a risk factor management evaluation database, a real-time judgment basis database, a judgment rule database and an accident case database, the cloud database (500) is used for storing risk factor information data acquired by the intelligent physical association and risk factor data acquisition system (100), the system is used for storing voice information contrastive analysis, keyword recognition, continuous voice recognition, a grammar analyzer, semantic information analysis and image characteristic information contrastive analysis of a risk factor management system (200), daily activity sign reasoning, weekly activity sign reasoning, monthly activity sign reasoning, quarterly activity sign reasoning, annual activity sign reasoning, early warning prompt information generation, early warning value evaluation, early warning level alarm information generation and early warning signal feedback to an upper-level early warning system, and updating information data in real time according to data transmitted by cloud computing (300) through cloud storage (400), the information data used for storing the prompt information, grading information, early warning information and alarm information generated by the artificial intelligent early warning operating system (600) and transmitted to a township artificial intelligent early warning system and the information used for storing information data generated by a multi-dimensional man-machine-object cooperative operation system (620), the system comprises a cloud database information search system, a four-level artificial intelligence early warning system (1000), an artificial intelligence early warning server (700), a cloud database information search system, a four-level artificial intelligence early warning system (1000), a three-level artificial intelligence early warning system (1100), a two-level artificial intelligence early warning system (1200), and a two-level artificial intelligence early warning server (700), wherein the cloud database information search system is used for searching cloud database information;

artificial intelligence early warning operating system (600): an AI artificial intelligence early warning operating system developed on the basis of a Linux operating system architecture comprises a brain-like neural network system (610), a multi-dimensional man-machine-object cooperative interoperation system (620), a public safety intelligent monitoring early warning and prevention control system (630), an autonomous unmanned servo system (640), a heaven-earth integrated information network platform system (650), a computer running program for managing and controlling computer hardware, software and data resources, interfaces for communicating each level of artificial intelligence early warning system with the Internet and a distributed early warning kiosk, interfaces for communicating cloud computing, cloud storage, a cloud database and the artificial intelligence early warning system, the Internet and the distributed early warning kiosk and other software, communication interfaces for the man-machine-object cooperative interoperation system, mobile equipment and an intelligent television and a human-machine interface for providing support for other application software, the system comprises a brain-like neural network system, a multi-dimensional man-machine-object cooperative interoperation system, a public safety intelligent monitoring, early warning, prevention and control system, an autonomous unmanned servo system, a heaven-earth integrated network information platform system, an intelligent internet of things and risk factor data acquisition system and a risk factor management system, wherein a subsystem of an artificial intelligent early warning operation system (600) comprises a voice recognition system, a machine vision system, an actuator system, a cognitive behavior system, a file system, process management, inter-process communication, memory management, network communication, a safety mechanism, a driving program and a user interface;

artificial intelligence early warning server (700): the system comprises a high-performance Central Processing Unit (CPU), an image processor (GPU), a programmable logic gate array (FPGA), a neural Network Processor (NPU), a heterogeneous/reconfigurable processor and an artificial intelligent early warning server (700), wherein the server is used for providing various high-performance computing services for a client in an early warning network system, and under the control of an artificial intelligent early warning operating system, the server provides services of centralized computing, information issuing and data management for a remote monitoring client through a network video server, an infrared night vision detection server, an infrared voice detection server, a video television wall server, a program controlled switch, an AI cloud computing server, an AI database server, a GPU cloud server, a Web server, a communication server, a display, a mixing matrix, a router and a modem which are connected with the server;

internet hand distributed early warning kiosk (800): namely an unattended early warning pavilion, which comprises a display, an artificial intelligent early warning server, a network video server, an infrared night vision detection server, an infrared voice detection server, an NVR network video recorder, a disk matrix, a video television wall server, a program controlled switch, a Web server, a communication server, a display, a high-definition network video decoder, a high-definition network digital matrix, a comprehensive management platform system, a streaming media server, an optical terminal, a router, a modem, a photovoltaic power generation module, a wireless charging receiver, a storage battery pack, a power distribution system, an air conditioning system, an access control system, a smoke sensing system, an infrared sensing system, an early warning factor control module, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module, an output early warning module and a wire mesh, and is used for placing hardware facilities of the artificial intelligent early warning system, the system comprises a face recognition system (110), a voice recognition system (120), a dynamic recognition system (130), a biological characteristic recognition system (140), an infrared voice detection system (150), an infrared night vision detection system (160), an early warning unmanned aerial vehicle cruise system (170) and risk factor information data collected by an intelligent Internet of things security monitoring system (180) for receiving monitoring points, outputting voice early warning alarm information and image early warning alarm information, and a five-level artificial intelligent early warning system (900) for consulting camera probes, voice monitors, infrared voice probes, infrared night vision probes and early warning unmanned aerial vehicle blanket type scanning information distributed at each monitoring point, receiving operation instructions sent by the five-level artificial intelligent early warning system and tracking abnormal target positioning sent to a space-ground integrated information network platform system by an early warning police kiosk, risk factor data information is collected, meanwhile, wide-area coverage is carried out on monitoring blind areas of the distribution control points, deep monitoring and global cooperation are carried out, safe and efficient real-time data transmission is carried out with a multi-dimensional man-machine object cooperative interoperation system (620), an auxiliary decision is provided for commanders, the early warning unmanned aerial vehicle can carry out real-time patrol on abnormal environments, and tracking and monitoring are carried out on abnormal behaviors;

five-level artificial intelligence early warning system (900): the five-level artificial intelligence early warning system belongs to a village and tranquility level early warning system and comprises a television wall display, an artificial intelligence early warning server, a network video storage server, a video television wall server, an NVR network video recorder, a magnetic disk matrix, a video terminal, a high-definition network video decoder, a decoder service switch, a program-controlled switch, a network facility cabinet, a multipoint control unit MCU, a Web server, a communication server, a display, a high-definition network digital matrix, an optical terminal, a comprehensive management platform system, a streaming media server, a router, a modem, a UPS uninterrupted power supply, a UPS storage battery pack, an air conditioning system, an early warning factor control module, an equipment access gateway, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module and an output early warning module, and is used for monitoring risk source early warning information of the surrounding environment of the Internet + distributed early warning kiosk (800) in, the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by an internet + distributed early warning police kiosk (800) and feeding back the early warning information to a four-stage artificial intelligent early warning system in time, is used for receiving an operation instruction sent by the four-stage artificial intelligent early warning system and replying an execution step, a process, a measure and an implementation result, is used for early warning an unmanned aerial vehicle to carry out real-time patrol on an abnormal environment and track and monitor abnormal behaviors, is used for a multi-dimensional man-machine-object cooperative interoperation system (620) to carry out data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization between the same IP address network segment and a mobile intelligent terminal device, is used for the early warning police kiosk to track the abnormal target position sent by a world integrated information network platform system, collects risk factor data information and covers a monitoring blind area of a distribution point, deep monitoring and global cooperation are realized, and safe and efficient real-time data transmission is realized through a system (620) which is interoperable with multi-dimensional man-machine objects in a cooperative manner, so that auxiliary decisions are provided for commanders;

four-level artificial intelligence early warning system (1000): the four-level artificial intelligence early warning system belongs to a county level early warning system and comprises a television wall display, an artificial intelligence early warning server, a network video storage server, a video television wall server, an NVR network video recorder, a disk matrix, a video terminal, a high-definition network video decoder, a decoder service switch, a program-controlled switch, a network facility cabinet, a multipoint control unit MCU, a Web server, a communication server, a display, a high-definition network digital matrix, an optical terminal, an integrated management platform system, a streaming media server, a router, a modem, a UPS uninterrupted power supply, a UPS storage battery pack, an air conditioning system, an early warning factor control module, an equipment access gateway, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module and an output early warning module, and is used for monitoring the Internet + distributed early warning pavilion (800) and the five-level artificial intelligence early warning system (900) in the jurisdiction and risk source early warning information The early warning information comprises audio data and/or video data, is used for receiving early warning information sent by an internet + distributed early warning kiosk (800) and a five-level artificial intelligent early warning system (900), timely feeds back the early warning information to the three-level artificial intelligent early warning system, is used for receiving an operation instruction sent by the three-level artificial intelligent early warning system and replying an execution step, a process, a measure and an implementation result, is used for early warning an unmanned aerial vehicle to carry out real-time patrol on abnormal environments and track and monitor abnormal behaviors, is used for a multi-dimensional man-machine-object cooperative interoperation system (620) to carry out data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization between the same IP address network segment and a mobile intelligent terminal device, is used for tracking abnormal target positioning sent by the early warning kiosk to a world integrated information network platform system and collecting risk factor data information, meanwhile, wide-area coverage is carried out on monitoring blind areas of the distribution control points, deep monitoring and global cooperation are carried out, and safe and efficient real-time data transmission is carried out with a multi-dimensional man-machine object cooperative interoperation system (620), so that an auxiliary decision is provided for commanders;

three-level artificial intelligence early warning system (1100): the three-level artificial intelligence early warning system belongs to a city level early warning system and comprises a video wall display, an artificial intelligence early warning server, a network video storage server, a video wall server, an NVR network video recorder, a disk matrix, a video terminal, a high-definition network video decoder, a decoder service switch, a program-controlled switch, a network facility cabinet, a multipoint control unit MCU, a Web server, a communication server, a display, a high-definition network digital matrix, an optical transceiver, an integrated management platform system, a streaming media server, a router, a modem, a UPS uninterrupted power supply, a UPS storage battery pack, an air conditioning system, an early warning factor control module, an equipment access gateway, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module and an output early warning module, wherein the artificial intelligence early warning system is used for monitoring the Internet + distributed early warning pavilion (800), The early warning system comprises a five-level artificial intelligence early warning system (900), a four-level artificial intelligence early warning system (1000) and risk source early warning information of the surrounding environment, wherein the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by the internet and a distributed early warning police kiosk (800), the five-level artificial intelligence early warning system (900) and the four-level artificial intelligence early warning system (1000), and timely feeds back the early warning information to the two-level artificial intelligence early warning system, is used for receiving an operating instruction sent by the two-level artificial intelligence early warning system, replying execution steps, processes, measures and implementation results, is used for early warning an unmanned aerial vehicle to carry out real-time patrol on abnormal environment, tracking and monitoring abnormal behaviors, and is used for carrying out data transmission, analysis, display and the like between the same IP address network segment and a mobile intelligent, UI switching, instruction execution, control and interoperation are synchronous, the system is used for tracking abnormal target positioning sent by an early warning police kiosk to a heaven-earth integrated information network platform system, acquiring risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a control point, carrying out deep monitoring and global cooperation, carrying out safe and efficient real-time data transmission with a multi-dimensional man-machine cooperative interoperation system (620), and providing an auxiliary decision for commanders;

a secondary artificial intelligence early warning system (1200): the second-level artificial intelligence early warning system belongs to a provincial hall level early warning system and comprises a video wall display, an artificial intelligence early warning server, a network video storage server, a video wall server, an NVR network video recorder, a disk matrix, a video terminal, a high-definition network video decoder, a decoder service switch, a program-controlled switch, a network facility cabinet, a multipoint control unit MCU, a Web server, a communication server, a display, a high-definition network digital matrix, an optical terminal, a comprehensive management platform system, a streaming media server, a router, a modem, a UPS uninterrupted power supply, a UPS storage battery pack, an air conditioning system, an early warning factor control module, an equipment access gateway, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module and an output early warning module, wherein the two-level artificial intelligence early warning system is used for monitoring the Internet + distributed early warning pavilion, The early warning system comprises a five-level artificial intelligence early warning system (900), a four-level artificial intelligence early warning system (1000), a three-level artificial intelligence early warning system (1100) and risk source early warning information of the surrounding environment, wherein the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by the internet + a distributed early warning kiosk (800), the five-level artificial intelligence early warning system (900), the four-level artificial intelligence early warning system (1000) and the three-level artificial intelligence early warning system (1100) and feeding back the early warning information to the one-level artificial intelligence early warning system (1300) in time, is used for receiving an operation instruction sent by the one-level artificial intelligence early warning system and replying execution steps, processes, measures and implementation results, is used for early warning an unmanned aerial vehicle to carry out real-time patrol the abnormal environment and track and monitor abnormal behaviors, and is used for carrying out data transmission between the same IP address network, The system comprises an analysis system, a display system, a UI switching system, an instruction execution system, a control system and an interoperation system, wherein the analysis system, the display system, the UI switching system, the instruction execution system, the control system and the interoperation system are synchronous, and are used for tracking abnormal target positioning sent to a heaven-earth integrated information network platform system by an early warning police kiosk, collecting risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a deployment and control point, carrying out deep layer monitoring and global cooperation, carrying out safe and efficient real-time data transmission with a multi-;

a first-level artificial intelligence early warning system (1300): the first-level artificial intelligence early warning system belongs to a department-level early warning system and comprises a video wall display, an artificial intelligence early warning server, a network video storage server, a video wall server, an NVR network video recorder, a disk matrix, a video terminal, a high-definition network video decoder, a decoder service switch, a program-controlled switch, a network facility cabinet, a multipoint control unit MCU, a Web server, a communication server, a display, a high-definition network digital matrix, an optical terminal, an integrated management platform system, a streaming media server, a router, a modem, a UPS uninterrupted power supply, a UPS storage battery pack, an air conditioning system, an early warning factor control module, an equipment access gateway, an optical fiber transceiver, a network adapter, a communication module, an input early warning factor module and an output early warning module, and is used for monitoring the Internet + distributed early warning pavilion (800) in the jurisdiction, A five-level artificial intelligence early warning system (900), a four-level artificial intelligence early warning system (1000), a three-level artificial intelligence early warning system (1100), a two-level artificial intelligence early warning system (1200) and risk source early warning information of the surrounding environment, wherein the early warning information comprises audio data and/or video data, is used for receiving early warning information sent by the Internet and a distributed early warning kiosk (800), the five-level artificial intelligence early warning system (900), the four-level artificial intelligence early warning system (1000), the three-level artificial intelligence early warning system (1100) and the two-level artificial intelligence early warning system (1200) and feeding the early warning information back to the two-level artificial intelligence early warning system (1200) in time, is used for sending an operation instruction to the two-level artificial intelligence early warning system and executing steps, processes and measures, is used for early warning an unmanned aerial vehicle to carry, the system is used for the multi-dimensional man-machine-object cooperative interoperation system (620) to carry out data transmission, analysis, display, UI switching, instruction execution, control and interoperation synchronization between the same IP address network segment and the mobile intelligent terminal device, is used for tracking abnormal target positioning sent out by an early warning police box to a heaven-earth integrated information network platform system, collecting risk factor data information, meanwhile, carrying out wide area coverage on a monitoring blind area of a control point, deep monitoring and global cooperation, safe and efficient and real-time data transmission with the multi-dimensional man-machine-object cooperative interoperation system (620), provides auxiliary decision for commanders, is used for vertical management, unified scheduling of the artificial intelligent early warning system and stably mastering the global for maintaining frontier.

2. The artificial intelligence early warning system of claim 1, wherein: the face recognition system (110) comprises face detection, face capturing and tracking, face comparison, face modeling and retrieval, wherein the face recognition system is used for extracting face feature points in the process of processing an image video by a computer, analyzing, identifying, modeling and retrieving by utilizing the principle of biometrics to find whether a specified person exists, the face detection comprises a reference template, face rules, sample learning, a skin color model and a feature sub-face, the face detection is used for judging whether a face to be searched exists in a dynamic scene or a complex environment and separating the face, the face capturing and tracking comprises face tracking, dynamic target tracking and skin color model tracking, the face capturing and tracking is used for detecting the face in one frame of an image or a video stream and separating the face from the background and automatically storing the face, and the face tracking is used for capturing and tracking the face, when the designated portrait moves in the range shot by the camera, the designated portrait is automatically tracked, the face comparison comprises face comparison, face identification, a characteristic vector and a face pattern template, the face comparison is used for the mode comparison of the face identification, the face identification is divided into two comparison modes of a verification mode and a search mode, the verification mode comparison is used for comparing and verifying the captured portrait or the designated portrait with a certain object registered in a database, the search mode comparison is used for searching and searching whether the designated portrait exists in all the registered portraits in the database, the face modeling and the search are used for modeling the portrait data registered in the database to extract the characteristics of the face, generating a face template for filing and face coding and storing the face template in the database, modeling the designated portrait when the face search is carried out, and then comparing and identifying the designated portrait with the templates of all persons in the database, the most similar people list will be listed according to the compared similarity values.

3. The artificial intelligence early warning system of claim 1, wherein: the voice recognition system (120) comprises a measurement and control computer, a controller, a voice recognition unit, a sound intensity detection unit, a voice synthesis unit, a panel control unit, an original voice comparison unit, a keyword recognition unit, a continuous voice recognition unit, a grammar analysis unit, a semantic information analysis unit, a voice information database and a voice coding and decoding record playback unit, and the voice recognition system (120) is used for realizing the operations of voice recognition, voice control, man-machine conversation and keyword dynamic editing, and is used for recognizing the voice of a specific person and recognizing multi-person voice.

4. The artificial intelligence early warning system of claim 1, wherein: the dynamic identification system (130) comprises information snapshot collection, dynamic portrait identification and gait identification, wherein the information snapshot collection is used for carrying out face detection and face extraction on pedestrians passing through the camera and storing and managing the information snapshot collection, after the system sets face detection rules for a channel, when the face in a picture reaches detection conditions, the system automatically shoots the face, then the captured qualified face image is sent to a background face comparison server for comparison and identification, the dynamic identification system (130) is used for carrying out dynamic portrait detection and image extraction on the pedestrians passing through the camera, people walk through an identification area in a natural form, the camera can carry out snapshot and collection of information and send corresponding instructions for carrying out dynamic face identification, the gait identification is used in a video intelligent monitoring occasion, and the system carries out an identity identification process on the walking posture of the pedestrians, and the gait identification system comprises a monitoring camera, The computer, gait video/image sequence processing system and recognition software are characterized in that a monitoring camera collects the gait of a person, a video sequence of the gait is obtained through detection and tracking, the gait feature of the person is extracted through preprocessing and analysis, the gait motion in the image sequence is subjected to the key steps of motion detection, motion segmentation and feature extraction gait recognition early stage, the gait feature is stored in a database in a gait mode through calculation processing, then the newly collected gait feature is compared and recognized with the gait feature of the gait database, if the gait feature is matched, the pre-alarming is carried out, and if the gait feature is not matched, the monitoring camera continues to carry out the gait collection.

5. The artificial intelligence early warning system of claim 1, wherein: the biological characteristic recognition system (140) comprises computer vision, image processing and model recognition, vehicle recognition, computer hearing, voice processing, a sensing network, fingerprint recognition of the Internet of things, face recognition, a skin chip, gait recognition, iris recognition, vein recognition, retina recognition, palm geometric recognition, DNA recognition, voice and handwriting recognition, paternity identification, hand shape recognition and signature recognition, the biological characteristic recognition system (140) is used for identifying the personal identity of the inherent physiological characteristics and behavior characteristics of a human body, firstly, the biological characteristics are sampled, the unique characteristics are extracted and converted into digital codes, the codes are further combined into characteristic templates, when people interact with the recognition system for identity authentication, the recognition system acquires the characteristics and compares the characteristics with the characteristic templates in a database to determine whether the codes are matched, thereby deciding to accept or reject the person.

6. The artificial intelligence early warning system of claim 1, wherein: the infrared voice detection system (150) comprises a transmitting device and a receiving device, the transmitting device comprises a laser and an optical collimating device, the receiving device comprises a photoelectric detector, an optical receiving system, a sound output device, an optical focusing device and an electric signal processing system, the infrared voice detection system (150) is used for remotely monitoring voice of a suspicious target room, firstly, a laser beam is used for irradiating the glass of the target room at a certain distance outside the target room, the reflected laser beam is modulated by a glass vibration signal caused by sound pressure, the reflected laser beam is received by the photoelectric sensor and converted into an electric signal, a circuit is used for filtering and amplifying the obtained signal, and meanwhile, audio processing software Adobe Audio filtering is used to cooperate with hardware filtering to optimize and restore the signal again, so that the monitoring of the target is realized.

7. The artificial intelligence early warning system of claim 1, wherein: the infrared night vision detection system (160) comprises an infrared night vision camera, a laser night vision infrared lamp, a CCD lens, a camera shooting main board and an infrared night video processor, wherein the infrared night vision detection system (160) is used for projecting infrared light to an object by adopting an infrared emitting device under a dark environment without visible light or low light, the infrared light enters the lens to be imaged after being reflected by the object, and the action trail of suspicious target personnel at night is monitored.

8. The artificial intelligence early warning system of claim 1, wherein: the early warning unmanned aerial vehicle cruise system (170) comprises a ground control center, an infrared optical antenna, a servo image stabilizing system, a multiband infrared search camera, an infrared tracking camera, a laser range finder, an early warning processing system, a tracking processing system, an information fusion processing system, a display and control terminal system, an inertial navigation system, a communication system, a data recording unit, a time sequence and power supply system, a micro control system, a power system, a remote controller, an attitude stabilizing system, an automatic navigation system and a target identification system, wherein the early warning unmanned aerial vehicle cruise system (170) is used for collecting high-resolution images, the unmanned aerial vehicle is provided with a high-definition digital video camera, a camera and a GPS positioning system, the unmanned aerial vehicle can perform positioning autonomous cruise, transmit shot images in real time, and monitoring personnel can synchronously watch and control the.

9. The artificial intelligence early warning system of claim 1, wherein: the intelligent Internet of things security monitoring system (180) comprises a sensing network, a transmission network, an application network, an Internet of things communication gateway, a video monitoring service subsystem, a data control module, a duplexer, a data transmitting module, a data receiving module, an encoder and a decoder, wherein the sensing network comprises intelligent household appliances, a monitoring terminal, an entrance guard terminal, an RF reader-writer, a Zigbee wireless data transmission network, Bluetooth, a GPRS general packet radio service network, Wi-fi, Rfid, Irda, Ipv6, an infrared sensor and an intelligent security Internet of things, the intelligent security Internet of things comprises a sensor node, a gateway, a router node and a security execution mechanism based on the wireless sensor network, the sensor node is used for acquiring Internet of things information, transmitting various security sensor data, generating an alarm signal after processing the data and informing a user of processing an emergency event in time, and simultaneously controlling the action of an actuating mechanism, wherein the router node is used for managing a network and is also used as a node of a sensor, distributing a network address for a child node of the router node and simultaneously helping other node routing data to send to a destination node, greatly enlarging the network range, the security actuating mechanism is used for the bottommost layer of the intelligent security Internet of things and is controlled by the sensor node and mainly responsible for executing and solving various security problems, the transmission network is used for transmitting a broadband network and a security Internet of things platform and transmitting the Internet of things of a telecommunication gateway and realizing the transmission and calculation of data, the application network comprises an Internet of things platform application, a gateway Internet of things application and a third party application consisting of an intelligent security monitoring platform and is used for realizing the tracking of a specific security management abnormal target, and the Internet of things communication gateway comprises an application system platform data interface module, a sensor, a, The system comprises a multimode network access processing module, a sensor data distribution module, an intranet processing module, an IP network layer interface, a serial port, a sensor network interface module and a USB peripheral equipment interface module, wherein an internet-of-things communication gateway is used for a monitoring center of the whole area network, can observe the running condition and alarm signals of the whole network in real time and realize interaction with a remote client through an internet network and a gsm network so as to inform a user of the occurrence of an alarm event, the user can also monitor the running condition of the whole network through the internet network and the gsm network and remotely control an executing mechanism, all monitoring data and alarm events are stored in a database server of the internet-of-things communication gateway, the application system platform data interface module is used for processing interfaces of an internet-of-things application platform and the gateway, and is used for sending and receiving sensing data, receiving instructions and sending return values to the internet processing module, the data channel is a data channel of the Internet of things and an Internet of things application platform, the sensing data distribution module is used for transferring and sending Internet of things data, the application system platform data interface module and the intranet processing module can register data in the sensing data distribution module, establish a data processing routing table of the data distribution module, determine the destination of data acquired from the sensing network according to the routing table sensing data distribution module, and simultaneously are responsible for sending data and instructions from the internal processing module to the sensing network, the intranet processing module is used for processing the application function of the gateway, receiving the data from the sensing network and making corresponding feedback according to the data to realize the closed-loop control function of the indoor Internet of things, the sensing network interface module is used for communicating with intelligent devices with various protocols, receiving the data generated by the devices and realizing data interaction between the sensing network devices and the gateway, the sensing network interface module has a uniform communication protocol format, the protocol adaptation between the sensing network interface module and various RF read-write equipment and intelligent equipment is completed through the equipment adaptation interface module, the video monitoring service subsystem is used for real-time interaction of remote audio and video, is used for completing compression coding of data by an audio terminal and a video terminal of a monitoring end, realizes encryption and encapsulation of network data through a specific protocol, realizes network transmission through the Internet, and is used for completing decoding and data synchronous processing of data by a client side, the state of a network is required to be detected in the process, meanwhile, a data compression mode is selected according to the state of the network so as to adapt to the unstable state of a network line, the data control module is used for receiving baseband data and generating digital signals from the baseband data, the duplexer is used for connecting an antenna device and receiving or sending modulation signals, the data transmitting module is used for connecting the data control module and transmitting a data signal to the duplexer, the data receiving module is used for connecting the data control module, receiving the data signal from the duplexer and transmitting the data signal to the data control module, the encoder is used for connecting the data transmitting module and generating a modulation signal with the data signal, and the decoder is used for connecting the data receiving module and generating the data signal with the received modulation signal.

10. The artificial intelligence early warning system of claim 1, wherein: the risk factor collecting module (210) comprises face collecting, voice collecting, pedestrian dynamic feature collecting and biological feature collecting, and the risk factor collecting module (210) is used for finding out predicted uncertain factors, analyzing the environmental condition and the sensitivity degree of the uncertain factors to an early warning scheme, evaluating related data, calculating the effect under various risk conditions and making correct judgment.

11. The artificial intelligence early warning system of claim 1, wherein: the risk factor identification module (220) includes steps of collecting information related to risk, determining risk factors, compiling risk identification reports, and the risk factor identification module (220) is used for identifying potential risk and behavior characteristics thereof, identifying main sources of risk and predicting possible consequences caused by the risk.

12. The artificial intelligence early warning system of claim 1, wherein: the risk factor reasoning module (230) is programmed by adopting a three-segment theory reasoning principle, wherein the three-segment theory reasoning principle comprises a big premise, a small premise and a conclusion, and the risk factor reasoning module (230) is used for finding out main risks faced by the security subject from an intricate and complex environment.

13. The artificial intelligence early warning system of claim 1, wherein: the risk factor evaluation module (240) mainly has the tasks of identifying various risks faced by an evaluation object, evaluating risk probability and possible negative effects, determining the risk bearing capacity of an organization, determining the priority of risk reduction and control, and recommending risk reduction countermeasures, and the risk factor evaluation module (240) is used for carrying out quantitative evaluation on the possibility that a possible safety event affects and loses various aspects of life, life and property of people before the risk event occurs, and measuring the possible degree of the effect or loss caused by a certain event or thing.

14. The artificial intelligence early warning system of claim 1, wherein: the risk factor response module (250) includes reacting to notifications of triggering events, executing risk action plans, reporting progress against plans, correcting off-plan conditions.

15. The artificial intelligence early warning system of claim 1, wherein: the risk management evaluation module (260) comprises a risk management environment, risk identification and reasoning, risk management control, risk management information exchange and feedback, risk management supervision and improvement, risk management and case and liability accident evaluation, and the risk management evaluation module (260) is used for analyzing, checking, correcting and evaluating the applicability and effectiveness of the risk management technology.

16. The artificial intelligence early warning system of claim 1, wherein: the brain-like neural network system (610) comprises a deep neural network model, a cognitive neural network model, a perception understanding decision neural network model, a neuron computing unit, a self-organizing unit, an association memory storage unit, a pattern recognition unit, a voice recognition unit, an information processing unit, an optimization combination unit, an intelligent control unit and a prediction evaluation unit, wherein the deep neural network model is used for deeply learning understanding of texts, images, audios, videos, gene data and complex network data which are available for massive large data, self-programming is completed through self-learning, the cognitive neural network model is used for converting visual, sensory, perceptual, memory, thinking, imagination, reasoning and language information into internal heart activity through information processing and outputting, and the perception understanding decision neural network model is used for environment perception of brain-like intelligent behaviors, The application of data understanding and reasoning decision-making capability can adapt to the environment, summarize the rule, complete operation, recognition or process control.

17. The artificial intelligence early warning system of claim 1, wherein: the multi-dimensional man-machine-object cooperative interoperation system (620): the media server authenticated by UPnP/DLNA is connected through bandwidth or WIFI network under the environment of cloud computing (300), cloud storage (400) and a cloud database (500), data transmission, analysis, display, UI switching, instruction execution and control and interoperation synchronization are carried out between different terminal devices, the mobile intelligent device and the intelligent television are accessed to the same local area network in the environment of cloud computing (300), cloud storage (400) and the cloud database (500), an IP address is set to be the same network segment, other devices are automatically discovered and connected through an interconnection mutual discovery module between the mobile intelligent device and the intelligent television, the mobile intelligent device is used for a message control mechanism, the interface is updated and switched, messages are forwarded and displayed, the instruction analysis and execution control and interoperation synchronization are carried out in the process of human-computer-object ternary interaction, and intelligent sharing of multimedia resources is completed.

18. The artificial intelligence early warning system of claim 1, wherein: the public safety intelligent monitoring early warning and prevention control system (630) comprises a public safety cloud computing solution, a crime information analysis solution, a crime prediction and prevention solution and an intelligent video monitoring solution, wherein the public safety cloud computing solution comprises a cloud computing management platform, a middleware cloud platform, a database cloud platform, an intelligent analysis platform, a virtualization technology platform and a storage virtualization technology platform, the public safety cloud computing solution is used for acquiring and processing safety information through the cloud computing information acquisition and processing cloud platform, the acquired object is aerospace acquired data based on an Internet of things platform, satellite remote sensing data, electromagnetic acquired data, video monitoring acquired data and night infrared detection acquired data, and real-time cleaning and processing integration of the safety information are realized through information fusion and the cloud computing processing platform, the method comprises the steps of realizing integration and ordered information access, management and sharing of safety information and GIS data high-value information data through information storage, realizing intelligent safety information visual display, lead decision auxiliary analysis and lead decision one-line implementation intelligent auxiliary decision through information and data analysis and prediction research and judgment, building a command cloud platform, realizing integration, real-time communication and command of headquarter commanders and one-line executives, and carrying out rapid evaluation and research and judgment on first-line battle fruits, wherein the crime information analysis solution comprises correlation analysis, network analysis, path analysis, time sequence analysis and space analysis, the correlation analysis in the crime information analysis solution is used for organically integrating different data from massive data to extract useful information, defining and finding correlation relation, and the network analysis is used for detecting all transmitted data in the network, Analyzing, diagnosing and analyzing paths for preventing crime, wherein time sequence analysis is used for mastering the time trend of criminals according to the time abnormality of crime incidence, time sequence analysis is used for defining the normal state of events and quantifying the deviation state of the normal state, space analysis is used for matching with the attribute information of space data, the potential information of space targets is mined through the joint analysis of the space data and space models, the basic information of the space targets is the space position, distribution, form, distance, direction and topological relation of the space targets, the distance, the direction and the topological relation form the space relation of the space targets, the space characteristics of geographic entities are used as the basis of data organization, query, analysis and reasoning, the morphological structures of different types of the targets are obtained by dividing the geographic space targets into different types of points, lines and planes, the method is characterized in that spatial data and attribute data of a spatial target are combined to perform spatial calculation and analysis of a given task, a crime prediction and prevention solution is used for identifying identities of persons, criminals, criminal suspects and victims who are in favorable relation to cases, is used for cooperating with other organizations and sharing data and information, is used for obtaining critical information related to cases and investigation and further defining crime modes for crime prediction, and is used for establishing effective police deployment with self-adaption capability and evaluating and commanding efficiency, an intelligent video monitoring solution consists of a front-end acquisition subsystem and a monitoring center, wherein the front-end acquisition subsystem comprises a network high-definition dome camera, a network high-definition infrared gun, an access switch, an optical fiber transceiver, a receiving optical terminal, a transmitting optical terminal and an ONU optical network unit, The system comprises an ODN optical distribution network, an OBD optical splitter, an EPON passive optical network, an OLT optical line terminal, a twisted-pair line transmitter and a transmission network subsystem, wherein a monitoring center comprises a back-end storage subsystem, an NVR network hard disk video recorder, a large display screen subsystem, a DVD player, a high-definition network video decoder, a high-definition network digital matrix, a comprehensive management platform system, a streaming media server, a disk array, a display host, a matrix control host, a monitoring keyboard, an Ethernet switch and client equipment, a network high-definition infrared gun is adopted at the front end and is transmitted through the Ethernet, digital video signals are directly input into the NVR for storing video images, the EPON is connected to the OLT through a single optical fiber, then the OLT is connected to the ONU, the IPTV, namely interactive network television and IAD voice comprehensive access equipment services are provided by the ONU, and three-in-one network application of voice, data and digital television, the front-end acquisition subsystem is used for acquiring data information of a video stream, an IPC (Internet control center) network camera with a certain function is selected according to different scenes to realize the optimal video monitoring effect, the IPC network camera is stored for 24 hours every day and stores storage spaces required by different days, an optical terminal is used for terminal equipment for optical signal transmission video, an ONU (optical network unit) is used for receiving broadcast data sent by an OLT (optical line terminal), an ODN (optical distribution network) is used for providing an optical transmission channel between the OLT and the ONU, an OBD (optical block device) splitter is used for separating out required resonance absorption lines, common light is divided into near ultraviolet rays, vacuum ultraviolet rays and extreme ultraviolet rays according to the wavelength after the light enters the splitter, an Ethernet passive optical network is used for point-to-multipoint passive optical transmission service, the OLT optical line terminal is used for connecting terminal equipment of an optical fiber trunk, the transmission network subsystem is used for accessing various monitoring resources, the monitoring center is used for deploying a rear-end storage subsystem, an NVR network video recorder, a large-screen display subsystem, a high-definition network video decoder, a high-definition network digital matrix, an integrated management platform system, a display host, a core switch and client equipment, wherein the rear-end storage subsystem is used for storing information acquired by a front end in real time, the storage period is 30 days, the storage equipment integrates video recording management, storage and streaming media forwarding functions, the platform and the client directly take streams from the storage equipment to preview and review, the large-screen display subsystem is used for displaying pictures acquired by the front end equipment, GIS system graphs, alarm information and other application software interfaces, and is also used for accessing local VGA signals, DVD signals and cable television signals, and the video integrated management platform can realize real-time preview, video splicing display, arbitrary segmentation, window-opening roaming, video-based display, Image superposition, image stretching and zooming, a high-definition network digital matrix server is deployed in a monitoring center for realizing simultaneous decoding and output of front-end multipath high-definition videos to a television wall and for finishing output, switching, storage, forwarding and remote control of the videos, realizing all switching functions of a traditional analog matrix and further realizing a video recording function of NVR, a VGA digital matrix is arbitrarily connected with any front-end camera channel in a system, is connected with video channel information of the system, acquires site information and camera position in real time, selects a video-recorded digital matrix according to requirements, remotely operates the digital matrix by using a mouse or a main control terminal or a management center, realizes automatic, manual switching, polling or group switching by a digital matrix decoding channel, displays a single picture or multiple pictures, supports automatic amplified display by alarming, and realizes a television wall solution with more channels by adopting a multi-machine seamless stacking mode in large-scale system application, video decoding previews images of different channels respectively, video automatic reconnection automatic recovery display is carried out after network disconnection recovery, a centralized storage video recording plan can be set to be circulated every week, every day and every hour, can be accurate to every minute and every second, video data are not lost after central power failure, the video recording plan is automatically executed after power recovery, images of an alarm site are forcibly switched to a specified window of a television wall during alarm, multi-channel simultaneous alarm time-round inspection display is realized, real-time video, video playback and grouping time-round inspection are realized by matching with a platform management system, under the condition of insufficient bandwidth, the code stream image quality is connected through a remote setting and adjusting any channel, the quality of front end browsing and video recording is ensured, an Inter gigabit network port uses an I/OAT technology and comprises a local area network, a 3G telephone line, a DDN, an ISDN, an E1, a VPN, an optical fiber network, a hard disk video recorder, a video server, million high definition cameras and full series network audio and video stream, by the mixed video decoding of many producer's equipment of central control go up the wall, adopt TCP/IP input and output, 1000M net gape, carry out multimachine networking full cross matrix through 1000M switch and switch, high definition network video decoder is used for realizing penetrating through various networks through the cloud, has multiple cell-phone system control and includes: iPhone, Windows, Mobile, BlackBerry, Symbian, Android, 3G dialing function, WIFI module extension, multiple WEB browsers including IE, Chrome, Firefox, Safari, 2 USB2.0 interfaces, stably realizing USB mouse, backup, recording, upgrade operation, TV, VGA and HDMI simultaneous output, VGA, HDMI full high definition 1080P display output, network service support DHCP, PPPOE, FTP, DNS, DDNS, NTP, UPNP, EMAIL, IP authority, IP search, alarm center, matched WEB, client, SDK, easily realizing interconnection, ARSP with domain name service function, remote monitoring one-key starting, streaming media server including sewise software system, live broadcast server software, on-demand server software, virtual live broadcast server software, shear server software, content management system, UMS/streaming media server software, RTSP/streaming media server transcoding system, streaming media server including RTP/RTP, MMS, RTMP, transmit the video file to the customer end, for the user watches online, the flow media server inputs the source and supports UDP, RTMP, HTTP: TS mainstream transmission protocol, output protocol supports Web application playing requirement and android system and ios system playing requirement, has time-shifting and time-shifting video downloading function, user links digital control function, provides secondary development interface, performs distributed deployment, separates input and output through internal and external network cards to ensure independence and reliability of input and distribution, on-demand server software supports mp4, flv, mov, TS, wmv, mkv and rmvb multiple types of files to upload, supports more than two modes including H.264/AAC coding, m3u8 output function, mobile terminal selects different code streams to be self-adapted to play according to bandwidth condition, performs play request authentication, video arbitrarily drags to play, cloud deployment, provides secondary development interface, configures external transcoding server, improves transcoding efficiency, stream media server is used for collecting, caching, scheduling and transmitting and playing stream media content, compressing continuous audio and video information and then placing the compressed information on a network server, wherein the comprehensive management platform system comprises a basic platform, a platform Service, a Service logic module and an application subsystem, the basic platform is used for shielding the difference of an operating system, a database, security encryption and the encapsulation of a multimedia protocol and improving the operation efficiency and system compatibility of upper-layer application, the platform Service is used for providing authentication management Service, decoding, splicing, control and display, computer signal upper wall, central management Service, streaming media forwarding Service, storage Service, general Service of alarm management and third-party system access Service, the Service logic module is used for providing video, intelligence, alarm, entrance guard and security monitoring Service interfaces through induction and encapsulation of common services, the application subsystem is used for using various services provided by the platform through a web Service interface and showing specific services to a final user, the management platform supports C/S client, B/S client and mobile phone client, the decoding and mosaic display is used for decoding the highest 256-path D1/128-path 720P/64-path 1080P code stream, the computer signal upper wall is used for displaying a host to realize the upper wall display by accessing a VGA input board of a video integrated management system, a non-compression mode is adopted when the host is on the wall, the real-time requirement of a client is met while the video upper wall display quality is ensured, the high-definition computer video upper wall requirement of the client is perfectly solved, the central management service is used for realizing the front-end and rear-end equipment management, the signal command forwarding control processing of each unit, the alarm information receiving and processing and the service support information management, and simultaneously provides the authentication and authorization service of the user and the application support of network equipment management, including configuration management, safety management, fault management and performance management, the system comprises a streaming media forwarding service, a storage service, a centralized storage management mode and a distributed storage mode, and further comprises a mass storage management mode and a distributed storage mode, audio and video data storage, quick retrieval, a backup strategy, a data automatic repair technology, namely data supplement, alarm centralized storage and important event centralized backup management.

19. The artificial intelligence early warning system of claim 1, wherein: the autonomous unmanned servo system (640), namely an unmanned aerial vehicle background control center, comprises an aircraft, a control station, a communication link and a payload, wherein the aircraft is divided into a flight platform, a power device, a navigation flight control system, an electrical system, a power supply system, an autopilot, an attitude measurement system, a servo steering engine and an engine, the control station is divided into a ground station system, a display system, an interface system, a control system and a software system, the communication link is divided into an image transmission system, a digital transmission system, an airborne system and a ground system, the payload is divided into a communication device, an image device and other devices, the artificial intelligent early warning system adopts a six-rotor unmanned aerial vehicle which is used for carrying out image acquisition and data transmission on camera monitoring blind areas, carrying out real-time patrol on abnormal environments, carrying out tracking monitoring on abnormal behaviors, carrying out identity tracking on suspicious personnel, tracking on places, The behavior tracking, the control station is a flying control center of an unmanned aerial vehicle on the ground, an airborne vehicle or a ship, the man-machine interaction, the data transmission and the synchronous watching and control are realized, the control station is used for returning information and images including load data, state information and position information of all subsystems on the aircraft through a downlink communication link, the transmission and the recovery are controlled simultaneously, the control station is communicated with the outside to finish the acquisition of weather information, the network information transmission among the systems, the receiving task and the information reporting, the navigation system is divided into an inertial navigation INS, a satellite navigation system including a GPS and a Beidou, combined navigation and backup navigation include radar tracking, radio tracking and direct estimation, the communication link is divided into an uplink link and a downlink link, the uplink link is used for sending and storing flight path data, a person sends a flying control command in real time in a loop and sends the control command to an airborne task load and accessory equipment, the automatic pilot comprises a built-in sensor, a three-axis angular rate gyro, a three-axis accelerometer, a three-axis magnetometer, a two-nozzle airspeed sensor, an air pressure altimeter, a GPS receiver, a temperature sensor, an input/output interface, a PWM input, a PWM output, a power supply monitoring, an ADC, a data exchange interface RS-232/RS-485, an external sensor, an airspeed altitude combined sensor, an ultrasonic altimeter, a PWM signal and discrete signal expander, a flight data recorder, an oil mass sensor, a GLONASS or a Beidou GPS receiver, the automatic pilot is used for automatically controlling a regulating device of an aircraft track, keeping the aircraft posture and completing a specified flight task, the automatic control system comprises an automatic takeoff, a hover, a flight and a landing, two CPU hot backup, automatic switching during the fault, a single fault of a sensor is allowed, accurate attitude and position estimation are kept after the fault, after the communication is interrupted, the airplane is controlled to continuously execute tasks or return automatically according to the setting, high integration integrated flight control, a gyroscope, an accelerometer, communication, a GPS, an altimeter, multi-channel control, 16-channel control output, a multi-servo steering engine or peripheral equipment can be controlled, unmanned aerial vehicle parameters are monitored in real time, intelligent alarm display is realized, three-dimensional flight path planning is supported, an automatic pilot ground control station is used for supporting the three-dimensional flight path planning, a reference point of a no-fly area is set, airplane parameters are monitored in real time, the intelligent alarm display is realized, flight data and video pictures are synchronously recorded, various maps and coordinate formats are supported, a user can define an operation interface, the operation is, the unmanned aerial vehicle supports network data distribution and network remote control, attitude measurement is used for measuring the angular velocity and the acceleration of an object in a three-dimensional space, one IMU comprises three single-axis accelerometers and three single-axis gyroscopes, the accelerometers detect the acceleration signals of the object in independent three axes of a carrier coordinate system, the gyroscopes detect the angular velocity signals of a carrier relative to a navigation coordinate system, the attitude of the object is calculated according to the signals, strapdown attitude calculation is performed, multi-sensor fusion and Kalman filtering are performed, a servo steering engine is used for a control circuit board to receive control signals from signal lines, a motor is controlled to rotate and drives a series of gear sets to be transmitted to an output steering wheel after being decelerated, an output shaft of the steering engine is connected with a position feedback potentiometer, the steering wheel drives the position feedback potentiometer while rotating, the potentiometer outputs a voltage signal to the control circuit board for feedback, the control circuit board determines the rotation direction and speed of the motor according to the position of the motor, so that the motor stops when reaching a target, the engine is divided into a four-stroke reciprocating internal combustion engine, a two-stroke reciprocating internal combustion engine, a rotary engine, a vortex engine, an electric motor and a ground station system and is used for monitoring flight attitude, height and speed information, monitoring the working states of the engine and a sensor, monitoring the electric quantity of an airborne power supply and the oil quantity of an oil tank, monitoring and planning the flight path of an aircraft, and recording flight state data and image data.

20. The artificial intelligence early warning system of claim 1, wherein: the space-ground integrated information network platform system (650) adopts a Beidou commercial system, namely a Beidou satellite navigation system, the Beidou system consists of three parts, namely a space section, a ground section and a user section, the Beidou system space section consists of three orbit satellites, namely a plurality of geostationary orbit satellites, an inclined geosynchronous orbit satellite and a medium-circle geosynchronous orbit satellite, and a hybrid navigation constellation, the Beidou system ground section comprises a master control station, a time synchronization/injection station and a monitoring station, the Beidou system user section comprises chips, modules and antennas which are compatible with other Beidou satellite navigation systems, as well as terminal products, application systems and application services, the Beidou-based public security information system realizes dynamic scheduling integrated command of police resources, improves sound speed and execution efficiency, and is used for public security services of anti-terrorism, stability maintenance, police and security, and the space-ground integrated information network platform system comprises a public security vehicle command scheduling system, The system comprises policemen field enforcement, emergency event information transmission, application of public security time service, vehicle autonomous navigation, vehicle tracking monitoring, a vehicle intelligent information system and vehicle networking application, wherein the emergency event information transmission uses a special short message function of Beidou, and the real-time high-precision navigation and positioning service of 1-2 meters, decimeter level and centimeter level is provided in a service area by using a Beidou/GNSS high-precision receiver and a ground reference station network and by using broadcasting means of satellites, mobile communication and digital broadcasting.

21. The artificial intelligence early warning system of claim 1, wherein: the electrified wire netting is installed in the range 5 meters away from the outside of the internet and the distributed early warning police kiosk (800) and used for preventing the human body from overstepping electrified obstacles, and the electrified wire netting comprises a power frequency power supply, a power supply lead, a wire netting, a grounding device and a signal device.